Relationship Between Total Serum Bilirubin Levels

and Carotid and Femoral Atherosclerosis

in Familial Dyslipidemia
Antonio J. Amor, Emilio Ortega, Verónica Perea, Montserrat Cofán, Aleix Sala-Vila,
Isabel Nuñez, Rosa Gilabert, Emilio Ros

Objective—Bilirubin is a potent antioxidant that has been inversely related to cardiovascular disease. There is little information
on serum total bilirubin (TB) in relation to atherosclerosis in familial dyslipidemia. We assessed the association between
TB and carotid and femoral atherosclerosis in this high-risk group.
Approach and Results—We evaluated 464 individuals with familial dyslipidemia (56% men; median age, 48 years), 322
with familial hypercholesterolemia, and 142 with familial combined hyperlipidemia. Carotid and femoral arteries were
imaged bilaterally with a standardized ultrasonographic protocol. Mean and maximum intima-media thickness and plaque
presence (≥1.2 mm) and height were recorded. Cross-sectional associations between TB and atherosclerosis variables were
investigated in multivariable-adjusted models, including lipid values and hypolipidemic drug use. Inflammatory markers
(C-reactive protein, total leukocyte count, and lipoprotein[a]) were also determined. Increasing TB levels were associated
with decreasing intima-media thickness of all carotid segments (P<0.05, all). TB also related to carotid plaque, present in
78% of individuals, and to plaque burden (≥3 plaques), with odds ratios (95% confidence interval) 0.59 (0.36–0.98) and
0.57 (0.34–0.96) for each increase of 0.5 mg in TB, respectively. Findings were confirmed in a validation cohort of 177
subjects with nonfamilial dyslipidemia. Only the familial combined hyperlipidemia group, with higher inflammation-
related markers, showed an inverse association between TB and femoral plaque height (β=−0.183; P=0.030).
Conclusions—TB was inversely and independently associated with carotid plaque burden in familial and nonfamilial
dyslipidemia. These findings support the use of TB as a biomarker of atherosclerosis in this high-risk group.
Visual Overview—An online visual overview is available for this article.   (Arterioscler Thromb Vasc Biol. 2017;37:
2356-2363. DOI: 10.1161/ATVBAHA.117.310071.)
Key Words: atherosclerosis ◼ bilirubin ◼ cross-sectional studies ◼ hyperlipidemia, familial combined
◼ hyperlipoproteinemia type II

A therosclerosis is a complex process in which inflamma-

tion is believed to play a central role, with oxidative stress
and DNA damage induced by oxidized low-density lipopro-
teins as crucial factors.1–3 In fact, inflammatory markers, such
as high-sensitivity C-reactive protein (hsCRP), fibrinogen,
complement C3, or white blood cell count (WBCC), are pro-
spectively associated with incident cardiovascular events.4–8
Conversely, antioxidants are thought to protect against athero-
sclerosis and coronary artery disease.9,10
Bilirubin—an end metabolic product of heme degrada-
tion—has demonstrated potent antioxidant and anti-inflam-
matory properties, both in in vitro and in vivo studies, being
particularly effective at suppressing low-density lipoprotein
cholesterol oxidation and inhibiting leukocyte migration,
consequently hampering the formation of atherosclerotic
plaques.11–13 In nonselected population, an inverse association
between serum total bilirubin (TB) and surrogate measures of
cardiovascular disease (CVD), such as intima-media thickness
(IMT),14,15 coronary artery calcium score, severity of coro-
nary atherosclerosis,16–18 arterial stiffness,19 and flow-mediated
vasodilation,20 has been demonstrated. Beyond this preclinical
data, serum TB concentrations have been shown to be inversely
associated with the risk of overt CVD in several studies.21–27
Information on the role of TB in atherosclerosis among
familial dyslipidemias is scarce. In fact, there is only 1
cross-sectional study that showed lower TB levels among
patients with familial hypercholesterolemia (FH) and prev-
alent CVD compared with those with no clinical athero-
sclerotic disease.28 To our knowledge, there are no studies
addressing the relationship between TB and preclinical

Received on: July 18, 2017; final version accepted on: October 9, 2017.
From the Endocrinology and Nutrition Service, Institut d’Investigacions Biomèdiques August Pi Sunyer, Hospital Clínic, Barcelona, Spain (A.J.A., E.O.,
M.C., A.S.-V., E.R.); Centro de Investigación Biomédica en Red Fisiopatología de la Obesidad y Nutrición, Instituto de Salud Carlos III, Spain (A.J.A.,
E.O., M.C., A.S.-V., E.R.); Service of Endocrinology, Hospital Universitari Mútua de Terrassa, Spain (V.P.); and Vascular Unit, Centre de Diagnòstic per
l’Imatge, Institut d’Investigacions Biomèdiques August Pi Sunyer, Hospital Clínic, Barcelona, Spain (I.N., R.G.).
The online-only Data Supplement is available with this article at http://atvb.ahajournals.org/lookup/suppl/doi:10.1161/ATVBAHA.117.310071/-/DC1.
Correspondence to Antonio J. Amor, MD, Endocrinology and Nutrition Service, Hospital Clínic, Villarroel 170, 08036 Barcelona, Spain. E-mail
ajamor@clinic.ub.es
© 2017 American Heart Association, Inc.
Arterioscler Thromb Vasc Biol is available at http://atvb.ahajournals.org DOI: 10.1161/ATVBAHA.117.310071

2356
 Amor et al   Bilirubin and Atherosclerosis in Dyslipidemia    2357
atherosclerosis in this group of individuals at high car-
diovascular risk. In the present study, we assessed cross-
sectional associations between TB and atherosclerosis, as
evaluated ultrasonographically by IMT and plaque in the
carotid and femoral territories, in a group of individuals
with familial dyslipidemia, namely FH with genetic diag-
nosis, FH without known genetic defects (named autoso-
mal dominant hypercholesterolemia [ADH]), and familial
combined hyperlipidemia (FCH).
Materials and Methods
Materials and Methods are available in the online-only Data
Supplement.
Results
Subjects’ Characteristics
The median age of the study subjects was 48 (25th–75th per-
centiles; 38–57) years, with a slightly higher prevalence of men
(56%). Type 2 diabetes mellitus was present in 4.5%, hyperten-
sion in 20.1% and previous CVD in 12.9%. Baseline character-
istics according to dyslipidemia diagnosis are shown in Table 1.
As expected, the FCH group had higher body mass index and
waist circumference than the other 2 groups (P<0.001 for all
comparisons). Regarding laboratory parameters, all lipid val-
ues except triglycerides were higher in the FH and ADH groups
than in FCH (P<0.05 for all comparisons). Conversely, inflam-
mation-related variables (hsCRP and WBCC) were higher in
FCH (P<0.05). FCH subjects also showed more advanced ath-
erosclerosis than the other 2 groups (all carotid IMT-related
variables and carotid and femoral plaque burden; P<0.05 for all
comparisons). Finally, individuals with previous CVD (n=60)
showed more advanced atherosclerosis (carotid plaque pres-
ence, 98.3% versus 57.3% and presence of ≥3 carotid plaques,
65.0% versus 23.8% for subjects with and without previous
CVD, respectively; P<0.001 for both).
Associations Between TB and Clinical
and Laboratory Variables
All but 16 participants with TB >1.2 mg/dL (the cutoff to
diagnose Gilbert disease in the absence of liver dysfunction)
had levels in the normal range. There were no differences in
TB between the different dyslipidemia subtypes (Table 1). TB
was higher in men than in women (0.7 [0.5–0.8] versus 0.5
[0.4–0.7] mg/dL, respectively; P<0.001) but was unrelated
to age (Tables I and II in the online-only Data Supplement).
TB related directly to waist circumference (r=0.117; P<0.05),
alcohol consumption (r=0.126; P<0.01), alanine amino trans-
ferase (r=0.157; P=0.001), and γ-glutamyl transpeptidase
(r=0.096; P<0.05) and inversely to inflammation-related
variables: hsCRP (r=−0.156; P=0.009), WBCC (r=−0.087;
P=0.061), and smoking habit (0.5 [0.4–0.7] and 0.7 [0.5–
0.8] for smokers versus nonsmokers, respectively; P<0.001;
Tables I and II in the online-only Data Supplement). No sig-
nificant relationships were found with some oxidative mark-
ers (r=−0.078, P=0.199 and r=−0.060, P=0.268 for oxidized
low-density lipoprotein and lipoprotein[a], respectively),
other lipid variables (except for high-density lipoprotein
cholesterol; r=−0.142; P=0.002), or markers of cholesterol
absorption or synthesis (serum noncholesterol sterol levels;
Table II in the online-only Data Supplement).
Associations Between TB and
IMT-Related Variables
In the entire study sample, we found an inverse associa-
tion between TB and age- and sex-adjusted common carotid
artery-IMT and carotid bulb-IMT (both mean and maximum
IMT), as well as maximum IMT of any carotid segment
(P<0.05 for all comparisons; Table III in the online-only Data
Supplement). When TB was divided into tertiles, the age- and
sex-adjusted IMT of all the carotid segments increased as TB
decreased (P for lineal trend <0.05 for all; Figure 1; Figure I
in the online-only Data Supplement). No associations were
found with femoral IMT. These relationships remained sta-
tistically significant after adjusting for other risk factors asso-
ciated with atherosclerosis and hypolipidemic drug treatment
(P<0.05 for all comparisons; Table III in the online-only Data
Supplement).
Because the FCH group differed from the FH and ADH
groups in anthropometric variables, inflammatory markers,
and atherosclerosis variables (Table 1), we examined the
associations between TB and IMT by dyslipidemia subtype.
Overall, and consistently, associations were stronger in the
FCH group (Table IV in the online-only Data Supplement).
Associations Between TB and Plaque Burden
In the entire cohort, TB was inversely related to carotid plaque
presence (odds ratio, 0.57 [0.37–0.89] for each increment of
0.5 mg/dL in TB; P=0.013; Figure 2) and with maximum
height of carotid plaques (β=−0.120; P=0.001) in age- and
sex-adjusted models. Regarding plaque burden, age- and sex-
adjusted TB gradually decreased as the number of carotid
plaques increased (0.72±0.02, 0.65±0.02, and 0.61±0.03 mg/
dL, for none, 1–2, and ≥3 plaques, respectively) and the num-
ber of territories with plaques (0.72±0.03, 0.68±0.02, and
0.64±0.02 mg/dL, for none, 1 territory [carotid or femoral],
and both territories; P for lineal trend <0.05 for both; Figure
3; Figure II in the online-only Data Supplement, respec-
tively). After adjusting for other risk factors and hypolipid-
emic drug treatment (fully adjusted models 1 and 2), presence
of ≥3 carotid plaques and maximum height of carotid
plaques remained inversely associated with TB (P<0.05 for
all comparisons; Table 2; Table V in the online-only Data
Supplement). The inverse association of TB with plaque bur-
den existed irrespective of prior CVD status (Table VI in the
online-only Data Supplement).
As with IMT-related variables, the FCH group showed the
strongest relationships between TB and plaque measurements.
Both in age- and sex-adjusted and in fully adjusted models, it
was the only group that showed statistically significant rela-
tionships between TB and plaque-related variables (carotid
plaque presence, ≥3 carotid plaques, and maximum height of
carotid plaques; P<0.05 for all; Figure 2; Table 2). FCH also
showed other relationships with TB not present in the other
groups (P<0.05 for interaction). In age- and sex-adjusted
models, TB was inversely related with the sum of maxi-
mum height of all carotid plaques (β=−0.234; P=0.004) and
2358   Arterioscler Thromb Vasc Biol   December 2017

Table 1.  Characteristics of Study Subjects

FH (n=240) ADH (n=82) FCH (n=142) P Value
Clinical characteristics
 Male sex, % 122 (51) 45 (55) 93 (66) 0.020
 Age, y 46 (33–56) 51 (40–60) 53 (46–58) <0.001
 Current smokers, % 53 (22) 18 (22) 43 (30) 0.165
 Premature familial CVD, % 83 (41) 24 (31) 46 (35) 0.220
 Alcohol consumption, g/d 3 (0–10) 5 (0–15) 5 (0–20) 0.005
 Type 2 diabetes mellitus, % 10 (4) 3 (4) 8 (6) 0.734
 Hypertension, % 31 (13) 19 (23) 43 (31) <0.001
 Personal history of CVD, % 33 (14) 8 (10) 19 (13) 0.637
 BMI, kg/m2 25.1 (22.4–27.8) 25.5 (23.5–28.4) 26.7 (24.7–29.2) <0.001
 Waist circumference, cm 90 (80–100) 92 (86–99) 98 (91–104) <0.001
 SBP, mm  Hg 120 (115–129) 126 (120–135) 130 (120–142) <0.001
 DBP, mm  Hg 74 (68–80) 80 (71–85) 80 (75–87) <0.001
 Statin score* 221 (40–500) 95 (0–368) 60 (0–224) <0.001
Lipid values, mg/dL
 Total cholesterol 359 (324–424) 328 (297–357) 298 (270–341) <0.001
 Triglycerides 112 (78–161) 127 (90–164) 263 (199–357) <0.001
 LDL cholesterol 286 (246–346) 244 (229–281) 200 (172–234) <0.001
 HDL cholesterol 49 (43–62) 52 (44–64) 44 (38–51) <0.001
 Apo AI 134 (119–154) 149 (130–163) 136 (123–153) 0.004
 Apo B 188 (162–219) 164 (151–191) 160 (141–181) <0.001
 Lipoprotein(a) 22 (10–52) 36 (15–81) 15 (6–53) 0.001
Other laboratory characteristics
 Creatinine, mg/dL 0.9 (0.8–1.0) 1.0 (0.9–1.1) 1.0 (0.9–1.1) 0.008
 hsCRP, mg/L 1.1 (0.5–2.3) 1.1 (0.7–2.6) 2.2 (0.8–3.4) 0.004
 WBCC (per mm3) 6500 (5415–7523) 6505 (5500–7625) 7300 (5900–8645) <0.001
 Glucose, mg/dL 91 (84–98) 93 (85–99) 98 (91–106) <0.001
 TB, mg/dL 0.6 (0.5–0.8) 0.6 (0.5–0.8) 0.6 (0.5–0.8) 0.889
 ALT, UI/L 24 (19–31) 26 (19–36) 29 (23–41) <0.001
 GGT, UI/L 18 (13–30) 22 (15–37) 30 (20–48) <0.001
 Oxidized LDL, U/L† 86.8 (64.6–113.5) 80.3 (60.9–101.5) 82.4 (62.0–102.2) 0.302
 Plasma phytosterols (μMol/mMol cholesterol)‡ 3.54 (2.25–4.91) 3.93 (3.02–5.37) 3.02 (2.17–3.98) 0.001
 Plasma lathosterol (μMol/mMol cholesterol)§ 0.99 (0.61–1.46) 1.23 (0.77–1.62) 1.43 (1.05–2.05) <0.001
Atherosclerosis variables
 Mean CCA-IMT, mm 0.64 (0.54–0.74) 0.66 (0.54–0.78) 0.67 (0.59–0.79) 0.020
 Mean bulb-IMT, mm 0.78 (0.62–1.03) 0.80 (0.65–1.10) 0.92 (0.75–1.13) 0.001
 Mean ICA-IMT, mm 0.65 (0.53–0.81) 0.72 (0.60–0.92) 0.74 (0.63–0.96) <0.001
 Carotid plaque 134 (56) 53 (65) 103 (73) 0.005
 Mean F-IMT, mm 0.78 (0.50–1.45) 0.88 (0.55–1.46) 0.91 (0.59–1.50) 0.057
 Femoral plaque 145 (60) 51 (62) 103 (75) 0.017
 Both carotid and femoral plaque 108 (45) 39 (48) 84 (61) 0.008
Data are shown as n (percentage) or median (Q1–Q3). P values for group comparisons are reported. ADH indicates autosomal dominant hypercholesterolemia; ALT,
alanine aminotransferase; Apo AI, apolipoprotein AI; Apo B, apolipoprotein B; BMI, body mass index; CCA, common carotid artery; CVD, cardiovascular disease; DBP,
diastolic blood pressure; F, femoral; FCH, familial combined hyperlipidemia; FH, familial hypercholesterolemia; GGT, γ-glutamyl transpeptidase; HDL, high-density
lipoprotein; hsCRP, high-sensitivity C-reactive protein; ICA, internal carotid artery; IMT, intima-media thickness; LDL, low-density lipoprotein; SBP, systolic blood pressure;
TB, total bilirubin; and WBCC, white blood cell count.
*Duration of treatment in years by average statin dose received standardized to simvastatin in milligram.
†Data available for n=85 FH, n=42 ADH, and n=77 FCH.
‡Sum of sitosterol+campesterol. Data available for n=269.
§Data available for n=269.
 Amor et al   Bilirubin and Atherosclerosis in Dyslipidemia    2359

the carotid and femoral territories were found in 61% and

Figure 1. Age- and sex-adjusted maximum intima-media thick-
ness (IMT) of any carotid segment according to bilirubin tertiles in
464 individuals with familial dyslipidemia. Total bilirubin (TB) ter-
tiles: T1, ≤0.5 mg/dL; T2, 0.6 to 0.7 mg/dL; and T3, ≥0.8 mg/dL.
maximum height of femoral plaques (β=−0.212; P=0.007),
without major changes in fully adjusted models (Table 2).
Validation Cohort
The validation cohort was made up of 177 individuals with
severe, nonfamilial dyslipidemia (polygenic hypercholesterol-
emia) assessed in our lipid clinic in the same way as individu-
als of the main cohort. Their median age was 53 (45–60) years,
and 51% were men. Type 2 diabetes mellitus was present in
3%, hypertension in 19%, and prior CVD in 6%. Plaques in
50%, respectively (35% had plaques in both territories). Other
characteristics are shown in Table VII in the online-only Data
Supplement. Similar to the main study cohort, TB was higher
in men than in women (P<0.001; Table VIII in the online-only
Data Supplement) but was not related to age (Table IX in the
online-only Data Supplement). TB also showed an inverse
association with inflammatory-related variables (namely
hsCRP; r=−0.259; P=0.005) and a direct association with liver
function tests (Table IX in the online-only Data Supplement).
Regarding relationships with atherosclerosis, TB was
inversely related to IMT in all carotid territories (particularly
in internal carotid artery, r=−0.149), but statistical significance
was not reached (Table X in the online-only Data Supplement).
In age- and sex-adjusted models, each increase of TB by 0.5
mg/dL showed a borderline significant inverse association
with carotid plaque presence (odds ratio, 0.49 [0.24–1.02];
P=0.056) and the sum of maximum heights of carotid
plaques (β=−0.143; P=0.054; Table XI in the online-only
Data Supplement), without major changes in fully adjusted
models (Table XII in the online-only Data Supplement). Like
in the main study cohort, age- and sex-adjusted TB gradu-
ally decreased as the number of carotid plaques increased
(0.76±0.04, 0.68±0.05, and 0.61±0.04 mg/dL, for none,
1, and ≥2 plaques, respectively; P=0.015 for lineal trend;
Figure III in the online-only Data Supplement). Similarly, TB
was inversely related to the presence of ≥2 plaques, both in
age- and sex-adjusted models (odds ratio, 0.41 [0.19–0.92];
P=0.032; Table XI in the online-only Data Supplement) and
in fully adjusted models (Table XII in the online-only Data

Figure 2. Age- and sex-adjusted models assessing the associations between TB and plaque-related variables in the total cohort of indi-
viduals with familial dyslipidemia and according to dyslipidemia subtype. Data expressed for an increase of 0.5 mg/dL in total bilirubin.
ADH indicates autosomal dominant hypercholesterolemia; CI, confidence interval; FCH, familial combined hyperlipidemia; FH, familial
hypercholesterolemia; and OR, odds ratio.
2360   Arterioscler Thromb Vasc Biol   December 2017

Table 2.  Associations Between Total Bilirubin and Plaque Burden Supplement). These relationships were stronger in individuals
in the Total Cohort of Individuals With Familial Dyslipidemia and with higher levels of inflammation markers (Table XIII in the
According to Dyslipidemia Subtype by Multiple Regression Analysis online-only Data Supplement). TB was unrelated to femoral
Carotid plaque presence (yes/no) plaques in this cohort.
Group OR (95% CI) P Value
 Total cohort 0.59 (0.36–0.98) 0.043 Discussion
In our study subjects with familial dyslipidemia, TB was
 FH 1.14 (0.53–2.47) 0.734
inversely associated with several measures of carotid and
 ADH 0.28 (0.02–4.13) 0.354 femoral atherosclerosis (both IMT and plaque-related vari-
 FCH 0.40 (0.18–0.89) 0.024 ables), independent of other classical and nonclassical cardio-
≥3 carotid plaques (yes/no) vascular risk factors or hypolipidemic drug treatment. Among
Group OR (95% CI) P Value
the dyslipidemia subtypes, FCH disclosed the highest levels
of systemic inflammation biomarkers (ie, hsCRP and WBCC)
 Total cohort 0.57 (0.34–0.96) 0.035
and the strongest associations between TB and plaque bur-
 FH 0.98 (0.42–2.29) 0.958 den. To the best of our knowledge, no prior study has related
 ADH 1.18 (0.05–27.14) 0.919 the serum levels of this powerful endogenous antioxidant to
 FCH 0.37 (0.15–0.96) 0.040 carotid and femoral atherosclerosis, including plaque-related
variables, in individuals at high cardiovascular risk because of
Plaques in carotid and femoral territories (yes/no)
familial dyslipidemia.
Group OR (95% CI) P Value
TB has previously been shown to be associated with
 Total cohort 0.70 (0.43–1.14) 0.154 reduced ultrasound-assessed carotid atherosclerosis in few,
 FH 1.17 (0.54–2.54) 0.691 usually small, studies.14,15,20,29,30 All but one15 used only com-
 ADH 0.38 (0.07–2.09) 0.269 mon carotid artery-IMT as outcome, and none assessed ath-
erosclerosis in the femoral arteries. Also, only 1 prior study
 FCH 0.52 (0.24–1.09) 0.085
conducted in elderly Japanese subjects showed an inverse
Maximum height of carotid plaque association between TB and carotid plaque29—a better inde-
Group β P Value pendent predictor of cardiovascular events than IMT31 recently
 Total cohort −0.089 0.015 included in the European Guidelines for CVD prevention as a
 FH −0.019 0.709 valid tool for cardiovascular risk reclassification.32 In the pres-
ent study, using a large number of individuals at high cardio-
 ADH −0.007 0.935
vascular risk and focusing on both carotid (including common
 FCH −0.214 0.006 carotid artery, bulb, and internal carotid artery) and femoral
Sum of maximum heights of all carotid plaques territories, we found that TB was inversely and independently
Group β P Value associated with both IMT and plaque-related variables in a
 Total cohort −0.061 0.121
dose-dependent manner. In fact, in multivariable-adjusted
models, an increase of 0.5 mg/dL in TB was associated with
 FH 0.026 0.620
a ≈50% chance of harboring a carotid plaque or ≥3 carotid
 ADH 0.018 0.848 plaques (Table 2; Figure 2). Importantly, statistical analyses
 FCH −0.196 0.019 considered the most known determinants of atherosclerosis
Maximum height of femoral plaque and statin treatment—a great modifier of the arterial wall fre-
quently overlooked in previous studies.
Group β P Value
Regarding the role of TB in CVD risk in familial dyslipid-
 Total cohort −0.045 0.257 emia, a single cross-sectional study in FH individuals showed
 FH 0.060 0.259 that TB was negatively associated with prevalent CVD in
 ADH −0.091 0.465 multiple regression models adjusted by common cardiovas-
 FCH −0.183 0.030 cular risk factors, but not by smoking status or previous statin
treatment, 2 variables closely linked to cardiovascular risk
Logistic regression models (OR and 95% confidence interval for an
increase in 0.5 mg/dL in total bilirubin) for dichotomous dependent variables and also related to TB levels in our cohort (Tables I and II
and lineal regression models (standardized regression coefficients) for log- in the online-only Data Supplement), which could limit the
transformed continuous dependent variables. ADH indicates autosomic findings of that study.28 In our study, focusing not only on
dominant hypercholesterolemia; CI, confidence iterval; FCH, familial combined genetically diagnosed FH, but also ADH and FCH, we found
hyperlipidemia; FH, familial hypercholesterolemia; HDL, high-density lipoprotein; an inverse association between this simple biomarker and ath-
LDL, low-density lipoprotein; and OR, odds ratio.
erosclerosis outcomes (Table 2; Tables III and V in the online-
Variables included in fully adjusted model 1 were age, sex, type of dyslipidemia
(FCH vs others), smoking habit, alcohol consumption, type 2 diabetes mellitus, only Data Supplement), even after adjusting for the most
hypertension, history of personal cardiovascular disease, body mass index, traditional cardiovascular risk factors (including smoking
statin score, actual treatment with bile acid sequestrants or ezetimibe, systolic habit and statin score), as well as the main variables driving
blood pressure, white blood cell count, γ-glutamyl transpeptidase, alanine the increased cardiovascular risk in this population, namely
aminotransferase, LDL cholesterol, HDL cholesterol, triglycerides, and lipoprotein(a). increased plasma lipid levels. In view of the recent interest
 Amor et al   Bilirubin and Atherosclerosis in Dyslipidemia    2361

Figure 3. Age- and sex-adjusted associations of total bilirubin with presence of plaques in ultrasound-assessed arterial territories (carotid
or femoral) in the total cohort of individuals with familial dyslipidemia (top) and by dyslipidemia subtype (bottom). ADH indicates autoso-
mal dominant hypercholesterolemia; FCH, familial combined hyperlipidemia; and FH, familial hypercholesterolemia.

in stratifying cardiovascular risk in familial dyslipidemia to
better tailor antiatherosclerotic treatment,33,34 the use of this
simple and widely available biochemical measure could serve
this purpose.
Among dyslipidemia subtypes, FCH showed the stron-
gest associations between TB and preclinical atherosclerosis.
Only in FCH did TB show statistically significant associations
with internal carotid artery-IMT in age- and sex-adjusted
models (P=0.077 for interaction between TB *dyslipidemia
group in Max IMT-internal carotid artery)—a carotid segment
where increased IMT is thought to best predict CVD events.35
FCH subjects showed increased inflammatory markers com-
pared with FH and ADH, and in them, TB also seemed to
confer a greater protection against plaque burden. A poten-
tial mechanism is an anti-inflammatory effect of bilirubin.
Concurring with our data (Tables I and II in the online-only
Data Supplement), a large population study showed an inverse
association between TB and markers of systemic inflamma-
tion (specifically, WBCC).36 Our findings, and those of oth-
ers showing an inverse relationship between TB and measures
of inflammation in patients with coronary atherosclerosis (ie,
hsCRP, neutrophil-to-lymphocyte ratio, and cell distribution
width),17 suggest that besides its antioxidant effects, TB may
also exert an anti-inflammatory action. FCH—a complex lipid
disorder in which inflammation and oxidative stress may play
a role in the lipid phenotype and attendant high cardiovascu-
lar risk37—could be particularly susceptible to the beneficial
effect of TB.
Higher levels of TB could relate to an impairment of the
enterohepatic circulation of cholesterol, leading to less bili-
ary cholesterol excretion, hence reduced intestinal cholesterol
absorption and ensuing lower blood cholesterol levels,38 thus
explaining, in part, the inverse association between TB and
atherosclerosis. However, on one hand, TB and total or low-
density lipoprotein cholesterol levels were unrelated, and on
the other hand, plasma levels of phytosterols (well-validated
markers for the efficiency of cholesterol absorption39) and
lathosterol (marker of cholesterol synthesis), determined in
a subsample of our cohort, also showed no association with
TB (Table II in the online-only Data Supplement), suggest-
ing that putative alterations of the enterohepatic circulation
of cholesterol played no significant role in our findings. The
strength of the genetic, cholesterol-raising defect in our study
subjects probably blunted any effect of TB in reducing blood
cholesterol.
Our study has some limitations. First, because of the cross-
sectional design, a causal relationship between TB and IMT or
plaque variables cannot be established. It is possible that TB
2362   Arterioscler Thromb Vasc Biol   December 2017
levels could decrease because of increased consumption dur-
ing enhanced oxidative stress, rather than being directly associ-
ated with atherosclerosis.40 Second, we have no information on
aspirin use in the whole cohort—a potential confounding factor
since it has been suggested that this drug raises bilirubin levels
through activation of heme oxygenase-1.41 Aspirin was used
by patients with CVD, who had increased IMT and plaques
more frequently than individuals in primary prevention, which
actually suggests an underestimation of the association. Third,
estrogen hormone replacement therapy can influence both lip-
ids and TB levels but, as customary in Spain,42 the proportion
of women treated with estrogens in our cohort was negligible.
Finally, seasonal variations in circulating bilirubin could also
influence the results43; however, ultrasonography and blood
tests were performed throughout the whole year, hence sea-
sonal variation probably played a minor role.
The study also has strengths. One is the relatively large
number of individuals with familial dyslipidemia evaluated
because there are no previous studies in this high cardiovascu-
lar risk population. Most prior studies assessing associations
of TB with atherosclerosis focused on IMT,14,15,20,30 whereas
we evaluated carotid plaque—a better independent predictor
of cardiovascular events.31 Furthermore, we adjusted our anal-
yses for all known variables that could influence atherosclero-
sis, including the statin score—a standardized measure of the
strength and duration of statin use frequently overlooked in
other studies. Finally, the main associations between TB and
carotid atherosclerosis were reproduced in a validation cohort
of individuals with nonfamilial (polygenic) dyslipidemia,
attesting to the robustness of our findings.
In conclusion, in a cross-sectional study, TB was inversely
associated with early (IMT) and advanced (plaque) mea-
surements of atherosclerosis in a population of familial
dyslipidemia, independent of other classical or nonclassical
cardiovascular risk factors, lipid profile, and statin treatment.
This finding was specially pronounced in FCH, in which
inflammation and oxidative stress may play an important
role in accelerated atherosclerosis. The inverse associations
between TB and carotid atherosclerosis were reproduced in
a validation cohort of individuals with polygenic hypercho-
lesterolemia. We hypothesize that high serum bilirubin, even
in the upper limit of physiological levels, might protect, in
part, dyslipidemic individuals from atherosclerosis. Our find-
ings also support using TB as an easy-to-measure, low-cost
biomarker of cardiovascular risk in this population, although
further studies with a larger sample and a longitudinal design
are needed to confirm our results.
Acknowledgments
CIBEROBN is an initiative of ISCIII, Spain.
Disclosures
E. Ros has received research funding from the California Walnut
Commission, Sacramento, CA, and is a nonpaid member of its
Scientific Advisory Committee. The other authors report no conflicts.
References
1. Libby P. Inflammation in atherosclerosis. Nature. 2002;420:868–874. doi:
10.1038/nature01323.
2. Stocker R, Keaney JF Jr. Role of oxidative modifications in atherosclero-
sis. Physiol Rev. 2004;84:1381–1478. doi: 10.1152/physrev.00047.2003.
3. Gisterå A, Hansson GK. The immunology of atherosclerosis. Nat Rev
Nephrol. 2017;13:368–380. doi: 10.1038/nrneph.2017.51.
4. Brown DW, Giles WH, Croft JB. White blood cell count: an independent
predictor of coronary heart disease mortality among a national cohort. J
Clin Epidemiol. 2001;54:316–322.
5. Ernst E, Resch KL. Fibrinogen as a cardiovascular risk factor: a meta-
analysis and review of the literature. Ann Intern Med. 1993;118:956–963.
6. Kannel WB, Anderson K, Wilson PW. White blood cell count and car-
diovascular disease. Insights from the Framingham Study. JAMA.
1992;267:1253–1256.
7. Muscari A, Bozzoli C, Puddu GM, Sangiorgi Z, Dormi A, Rovinetti C,
Descovich GC, Puddu P. Association of serum C3 levels with the risk
of myocardial infarction. Am J Med. 1995;98:357–364. doi: 10.1016/
S0002-9343(99)80314-3.
8. Ridker PM, Cushman M, Stampfer MJ, Tracy RP, Hennekens CH.
Inflammation, aspirin, and the risk of cardiovascular disease in appar-
ently healthy men. N Engl J Med. 1997;336:973–979. doi: 10.1056/
NEJM199704033361401.
9. Frankel EN, Kanner J, German JB, Parks E, Kinsella JE. Inhibition of
oxidation of human low-density lipoprotein by phenolic substances in red
wine. Lancet. 1993;341:454–457.
10. Ridker PM, Lüscher TF. Anti-inflammatory therapies for cardiovascular
disease. Eur Heart J. 2014;35:1782–1791. doi: 10.1093/eurheartj/ehu203.
11. Stocker R, Yamamoto Y, McDonagh AF, Glazer AN, Ames BN. Bilirubin
is an antioxidant of possible physiological importance. Science.
1987;235:1043–1046.
12. Wu TW, Fung KP, Wu J, Yang CC, Weisel RD. Antioxidation of human
low density lipoprotein by unconjugated and conjugated bilirubins.
Biochem Pharmacol. 1996;51:859–862.
13. Vogel ME, Idelman G, Konaniah ES, Zucker SD. Bilirubin prevents ath-
erosclerotic lesion formation in low‐density lipoprotein receptor‐deficient
mice by inhibiting endothelial VCAM‐1 and ICAM‐1 signaling. J Am
Heart Assoc. 2017;6:e004820. doi: 10.1161/JAHA.116.004820.
14. Vítek L, Novotný L, Sperl M, Holaj R, Spácil J. The inverse association
of elevated serum bilirubin levels with subclinical carotid atherosclerosis.
Cerebrovasc Dis. 2006;21:408–414. doi: 10.1159/000091966.
15. Dullaart RP, Kappelle PJ, de Vries R. Lower carotid intima media thick-
ness is predicted by higher serum bilirubin in both non-diabetic and type
2 diabetic subjects. Clin Chim Acta. 2012;414:161–165. doi: 10.1016/j.
cca.2012.08.029.
16. Tanaka M, Fukui M, Tomiyasu K, Akabame S, Nakano K, Hasegawa G,
Oda Y, Nakamura N. Low serum bilirubin concentration is associated with
coronary artery calcification (CAC). Atherosclerosis. 2009;206:287–291.
doi: 10.1016/j.atherosclerosis.2009.02.010.
17. Akboga MK, Canpolat U, Sahinarslan A, Alsancak Y, Nurkoc S,
Aras D, Aydogdu S, Abaci A. Association of serum total bilirubin
level with severity of coronary atherosclerosis is linked to systemic
inflammation. Atherosclerosis. 2015;240:110–114. doi: 10.1016/j.
atherosclerosis.2015.02.051.
18. Kang SJ, Kim D, Park HE, Chung GE, Choi SH, Choi SY, Lee W, Kim
JS, Cho SH. Elevated serum bilirubin levels are inversely associated with
coronary artery atherosclerosis. Atherosclerosis. 2013;230:242–248. doi:
10.1016/j.atherosclerosis.2013.06.021.
19. Li Y, Meng SY, Meng CC, Yu WG, Wang RT. Decreased serum bilirubin
is associated with arterial stiffness in men. Nutr Metab Cardiovasc Dis.
2013;23:375–381. doi: 10.1016/j.numecd.2011.09.004.
20. Erdogan D, Gullu H, Yildirim E, Tok D, Kirbas I, Ciftci O, Baycan
ST, Muderrisoglu H. 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;184:431–437. doi: 10.1016/j.atherosclerosis.2005.05.011.
21. Schwertner HA, Vítek L. Gilbert syndrome, UGT1A1*28 allele, and
cardiovascular disease risk: possible protective effects and therapeutic
applications of bilirubin. Atherosclerosis. 2008;198:1–11. doi: 10.1016/j.
atherosclerosis.2008.01.001.
22. Hopkins PN, Wu LL, Hunt SC, James BC, Vincent GM, Williams RR.
Higher serum bilirubin is associated with decreased risk for early familial
coronary artery disease. Arterioscler Thromb Vasc Biol. 1996;16:250–255.
23. Djoussé L, Levy D, Cupples LA, Evans JC, D’Agostino RB, Ellison RC.
Total serum bilirubin and risk of cardiovascular disease in the Framingham
offspring study. Am J Cardiol. 2001;87:1196–1200; A4, 7.
24. Troughton JA, Woodside JV, Young IS, Arveiler D, Amouyel P,
Ferrières J, Ducimetière P, Patterson CC, Kee F, Yarnell JW, Evans A;
 Amor et al   Bilirubin and Atherosclerosis in Dyslipidemia    2363

PRIME Study Group. Bilirubin and coronary heart disease risk in the
Prospective Epidemiological Study of Myocardial Infarction (PRIME).
Eur J Cardiovasc Prev Rehabil. 2007;14:79–84. doi: 10.1097/01.
hjr.0000230097.81202.9f.
25. Hunt SC, Wu LL, Hopkins PN, Williams RR. Evidence for a major gene
elevating serum bilirubin concentration in Utah pedigrees. Arterioscler
Thromb Vasc Biol. 1996;16:912–917.
26. Perlstein TS, Pande RL, Beckman JA, Creager MA. Serum total bili-
rubin level and prevalent lower-extremity peripheral arterial disease:
National Health and Nutrition Examination Survey (NHANES) 1999 to
2004. Arterioscler Thromb Vasc Biol. 2008;28:166–172. doi: 10.1161/
ATVBAHA.107.153262.
27. Horsfall LJ, Nazareth I, Petersen I. Cardiovascular events as a function
of serum bilirubin levels in a large, statin-treated cohort. Circulation.
2012;126:2556–2564. doi: 10.1161/CIRCULATIONAHA.112.114066.
28. de Sauvage Nolting PR, Kusters DM, Hutten BA, Kastelein JJ; EXPRESS
Study Group. Serum bilirubin levels in familial hypercholesterolemia: a
new risk marker for cardiovascular disease? J Lipid Res. 2011;52:1755–
1759. doi: 10.1194/jlr.P013193.
29. Kawamoto R, Ninomiya D, Hasegawa Y, Kasai Y, Kusunoki T, Ohtsuka N,
Kumagi T, Abe M. Mildly elevated serum bilirubin levels are negatively
associated with carotid atherosclerosis among elderly persons. PLoS One.
2014;9:e114281. doi: 10.1371/journal.pone.0114281.
30. Hamur H, Duman H, Demirtas L, Bakirci EM, Durakoglugil ME,
Degirmenci H, Kalkan K, Yildirim E, Vuruskan E. Total bilirubin levels
predict subclinical atherosclerosis in patients with prediabetes. Angiology.
2016;67:909–915. doi: 10.1177/0003319716632394.
31. Brook RD, Bard RL, Patel S, Rubenfire M, Clarke NS, Kazerooni EA,
Wakefield TW, Henke PK, Eagle KA. A negative carotid plaque area
test is superior to other noninvasive atherosclerosis studies for reduc-
ing the likelihood of having underlying significant coronary artery dis-
ease. Arterioscler Thromb Vasc Biol. 2006;26:656–662. doi: 10.1161/01.
ATV.0000200079.18690.60.
32. Piepoli MF, Hoes AW, Agewall S, et al; Authors/Task Force Members.
2016 European Guidelines on cardiovascular disease prevention in
clinical practice: the Sixth Joint Task Force of the European Society of
Cardiology and Other Societies on Cardiovascular Disease Prevention in
Clinical Practice (constituted by representatives of 10 societies and by
invited experts) developed with the special contribution of the European
Association for Cardiovascular Prevention & Rehabilitation (EACPR).
Eur Heart J. 2016;37:2315–2381. doi: 10.1093/eurheartj/ehw106.
33. Paquette M, Dufour R, Baass A. The Montreal-FH-SCORE: a new score
to predict cardiovascular events in familial hypercholesterolemia. J Clin
Lipidol. 2017;11:80–86. doi: 10.1016/j.jacl.2016.10.004.
34. Pérez de Isla L, Alonso R, Mata N, et al. Predicting cardiovascular events
in familial hypercholesterolemia: the SAFEHEART registry. Circulation.
2017;135:2133–2144. doi: 10.1161/CIRCULATIONAHA.116.024541.
35. Polak JF, Pencina MJ, Meisner A, Pencina KM, Brown LS, Wolf PA,
D’Agostino RB Sr. Associations of carotid artery intima-media thickness
(IMT) with risk factors and prevalent cardiovascular disease: comparison
of mean common carotid artery IMT with maximum internal carotid artery
IMT. J Ultrasound Med. 2010;29:1759–1768.
36. Tsai WN, Wang YY, Liang JT, Lin SY, Sheu WH, Chang WD. Serum total
bilirubin concentrations are inversely associated with total white blood
cell counts in an adult population. Ann Clin Biochem. 2015;52(pt 2):251–
258. doi: 10.1177/0004563214541969.
37. Martinez-Hervas S, Fandos M, Real JT, Espinosa O, Chaves FJ, Saez
GT, Salvador A, Cerdá C, Carmena R, Ascaso JF. Insulin resistance and
oxidative stress in familial combined hyperlipidemia. Atherosclerosis.
2008;199:384–389. doi: 10.1016/j.atherosclerosis.2007.11.023.
38. Bulmer AC, Verkade HJ, Wagner KH. Bilirubin and beyond: a review
of lipid status in Gilbert’s syndrome and its relevance to cardiovascular
disease protection. Prog Lipid Res. 2013;52:193–205. doi: 10.1016/j.
plipres.2012.11.001.
39. Escurriol V, Cofán M, Moreno-Iribas C, Larrañaga N, Martínez C, Navarro
C, Rodríguez L, González CA, Corella D, Ros E. Phytosterol plasma con-
centrations and coronary heart disease in the prospective Spanish EPIC
cohort. J Lipid Res. 2010;51:618–624. doi: 10.1194/jlr.P000471.
40. Breimer LH, Wannamethee G, Ebrahim S, Shaper AG. Serum bilirubin
and risk of ischemic heart disease in middle-aged British men. Clin Chem.
1995;41:1504–1508.
41. Grosser N, Abate A, Oberle S, Vreman HJ, Dennery PA, Becker JC,
Pohle T, Seidman DS, Schröder H. Heme oxygenase-1 induction may
explain the antioxidant profile of aspirin. Biochem Biophys Res Commun.
2003;308:956–960.
42. Baladé Martínez L, Montero Corominas D, Macías Saint-Gerons D.
Utilization of hormone replacement therapy in Spain: trends in the
period 2000-2014. Med Clin (Barc). 2016;147:287–292. doi: 10.1016/j.
medcli.2016.05.023.
43. Miyake K, Miyake N, Kondo S, Tabe Y, Ohsaka A, Miida T. Seasonal vari-
ation in liver function tests: a time-series analysis of outpatient data. Ann
Clin Biochem. 2009;46(pt 5):377–384. doi: 10.1258/acb.2009.008203.

Highlights
• Total bilirubin was inversely associated with carotid atherosclerosis in familial hypercholesterolemia with and without genetic defects and with
carotid and femoral atherosclerosis in familial combined hyperlipidemia.
• An increase of 0.5 mg/dL in total bilirubin was associated with roughly a 50% chance of having carotid plaques.
• In familial combined hyperlipidemia—a dyslipidemia phenotype with higher inflammation markers—total bilirubin shows the strongest inverse
associations with atherosclerotic variables.
• Total bilirubin might be used as a simple marker to improve cardiovascular risk stratification in familial dyslipidemia.