- Email: [email protected]
Elevated serum bilirubin levels are inversely associated with coronary artery atherosclerosis
Atherosclerosis 230 (2013) 242e248
Contents lists available at ScienceDirect
Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis
Elevated serum bilirubin levels are inversely associated with coronary artery atherosclerosis Seung Joo Kang a, Donghee Kim a, *, Hyo Eun Park a, Goh Eun Chung a, Seung Ho Choi a, Su-Yeon Choi a, Whal Lee b, Joo Sung Kim a, Sang-Heon Cho a a
Department of Internal Medicine, Healthcare Research Institute, Seoul National University, Hospital Healthcare System Gangnam Center, Seoul, South Korea Department of Radiology, Seoul National University College of Medicine, Seoul, South Korea
b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 September 2012 Received in revised form 20 June 2013 Accepted 22 June 2013 Available online 20 July 2013
Background: Inverse correlations of high serum bilirubin with metabolic and cardiovascular disease have been suggested. However, anti-atherogenic effects of bilirubin have not been well-established in terms of the presence of plaques and stenosis identified in coronary computed tomography (CT). Methods: A cross-sectional study was conducted on 2862 men who were free of cardiovascular disease and underwent coronary CT as part of a routine medical screening examination. Coronary stenotic lesions were considered to be incidences of coronary atherosclerosis, and stenosis was classified as stenosis <50% or 50%, according to degree of stenosis. Results: The prevalences of coronary atherosclerosis and stenosis 50% in subjects with elevated bilirubin levels (>1.2 mg/dL) were lower than those in subjects with normal bilirubin levels (1.2 mg/dL) (19.9% vs. 27.9%, p < 0.001, 8.5% vs. 10.3%, p ¼ 0.044). Bilirubin was inversely associated with total plaques (odds ratio [OR] 0.59, 95% confidence interval [CI] 0.48e0.73 in the 4th quartile vs. 1st quartile) and calcified plaques (OR 0.60, 95% CI 0.49e0.75) in univariate analysis. After adjusting for traditional risk factors, it was found that coronary atherosclerosis (OR 0.73, 95% CI 0.56e0.94 in the 4th quartile vs. 1st quartile) and calcified plaque (OR 0.66, 95% CI 0.53e0.84) were inversely associated with the bilirubin grade in a dose-dependent manner. Conclusions: The serum bilirubin level was inversely associated with coronary atherosclerosis and calcified plaques in a dose-dependent manner. These results suggested that serum bilirubin could be used as a protective biomarker of coronary artery disease. Ó 2013 Elsevier Ireland Ltd. All rights reserved.
Keywords: Bilirubin Atherosclerosis Coronary stenosis Plaque Atherosclerotic
1. Introduction Cardiovascular disease is the most common cause of mortality in developed countries and accounts for up to one-third of all deaths worldwide [1]. The overproduction of oxygen free radicals from oxidative stress is known to mediate various signaling pathways that underlie vascular inflammation in atherogenesis, from the
Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; CAC, coronary artery calcification; CT, computed tomography; CI, confidence interval; GGT, gamma-glutamyl transpeptidase; HDL, high-density lipoprotein; hsCRP, high sensitive c-reactive protein; LDL, low-density lipoprotein; OR, odds ratio; TG, triglycerides. * Corresponding author. Department of Internal Medicine, Healthcare Research Institute, Seoul National University Hospital, Healthcare System Gangnam Center, 39FL., Gangnam Finance Center 737, Yeoksam-Dong, Gangnam-Gu, Seoul 135-984, South Korea. Tel.: þ82 02 2112 5574; fax: þ82 02 2112 5635. E-mail addresses: [email protected] (S.J. Kang), [email protected] (D. Kim). 0021-9150/$ e see front matter Ó 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.atherosclerosis.2013.06.021
initiation of fatty streak development to ultimate plaque rupture [2]. Although the pathogenesis of atherosclerosis has not been thoroughly investigated, oxidative stress and DNA damage induced by oxidized low-density lipoprotein (LDL) cholesterol and by dietinduced hypercholesterolemia contribute to the progression of atherosclerosis [3]. Therefore, antioxidants are thought to serve a protective role against atherosclerosis and coronary artery disease by preventing the oxidative modification of LDL cholesterol [4]. Bilirubin is a potent antioxidant under physiological conditions and suppresses the oxidation of lipids and lipoproteins. Thus, bilirubin has been demonstrated in vitro to prevent plaque formation and subsequent formation of atherosclerosis [5]. A prospective cohort study suggested that low levels of bilirubin are correlated to premature coronary artery disease [6]. Previous studies have demonstrated the inverse relationship between serum total bilirubin concentration to peripheral artery disease and carotid intimaemedia thickness [7,8]. A genetic association between genes that influence serum bilirubin concentration such as heme
S.J. Kang et al. / Atherosclerosis 230 (2013) 242e248
oxygenase and uridine diphosphate-glucuronosyl transferase and coronary artery disease has been addressed in several studies [9,10]. In recent studies, a high serum bilirubin level was inversely associated with the presence of coronary artery calcification (CAC) [11,12]. However, CAC is limited in its prediction of noncalcified plaques, and as a result, it fails to represent the entire spectrum of atherosclerotic plaques. CAC screening is also inadequate in evaluating the degree of stenosis caused by plaque [13]. Although individuals with low bilirubin levels have a higher prevalence of coronary artery disease, there are few studies on the effect of serum total bilirubin on coronary artery stenosis and plaques in large populations. Thus, we investigated the effect of serum bilirubin on both coronary artery stenosis and coronary plaques, as assessed by coronary computed tomography (CT), in large, apparently healthy Korean males. 2. Methods 2.1. Study population and study design We performed a cross-sectional study in subjects who underwent a comprehensive medical check-up at Seoul National University Hospital, Healthcare System Gangnam Center from July 2006 to March 2010. Some of the subjects had voluntarily paid for a general health check-up, while others were supported through their employer. During this period, 3319 male subjects underwent a coronary CT examination for screening. All coronary CT and clinical and laboratory assessments were performed on the same day or within 6 months of each other. Out of the 3319 subjects, we excluded 146 subjects who had a history of heart attack, coronary artery disease (including acute myocardial infarction), angina, or congestive heart failure. We also excluded 249 subjects with at least one potential cause of chronic liver disease: 182 with positive hepatitis B surface antigen, 55 with positive hepatitis C antibody, and 12 with a history of other hepatitis or liver disease (e.g., primary biliary cirrhosis, autoimmune hepatitis, Wilson’s disease, etc.). An additional 34 subjects with abnormal biliary tracts (as observed in abdominal ultrasonography) and 28 subjects (mean standard deviation of creatinine: 1.61 0.26 mg/dL) with abnormal renal function (creatinine level > 1.4 mg/dL) were also excluded. Finally a total of 2862 subjects were enrolled in this study. The Institutional Review Board of Seoul National University Hospital approved the study protocol, and the study was conducted in accordance with the ethical guidelines of the 1975 Declaration of Helsinki.
243
antibody to the hepatitis C virus. Venous blood samples were taken from all subjects before 10 AM, following a 12-h overnight fast. Hypertension was defined as having a systolic blood pressure 140 mmHg, a diastolic blood pressure 90 mmHg and/or current use of antihypertensive medications. Subjects with a fasting plasma glucose level of 126 mg/dL or currently on anti-diabetic treatments were defined as having diabetes mellitus [14]. 2.3. Measurement of coronary artery stenosis and plaque by coronary artery CT Coronary CT was performed using a 16-row multi-slice CT (Sensation 16, Siemens Medical Systems, Erlangen, Germany) as previously described [15]. Briefly, after obtaining a topogram of the chest, a calcium score scan and angiography were performed using the retrospective method with a tube voltage of 120 kV and a 110 effective mAs tube current with a 200 mm field of view and ECGgated dose modulation. Prior to the CT scan, the patient’s heart rate was measured. Patients with a heart rate higher than 60 beats per minute were given a 50e100 mg dose of metoprolol. All of the images were analyzed by a radiologist who was blinded to the clinical and laboratory results of the subjects. Each lesion was identified using a multi-planar reconstruction technique and maximum intensity projection of the short-axis, two- and fourchamber views. The degree of stenosis and the plaque characterization were measured from the coronary CT images with a dedicated computer 3D workstation (Rapidia; Infinitt, Seoul, Korea). Coronary artery stenosis was estimated from the contrast enhanced portion of the coronary lumen. The stenosis was semi-automatically traced at the maximal stenotic site and then compared with the mean value for the proximal and distal reference sites [16]. Subjects with any coronary stenotic lesions were classified as having coronary atherosclerosis. Stenosis was classified as <50% or 50% based on the degree of stenosis. Number of stenosis was assessed using the 15-segment model of American Heart Association [17]. Plaques were identified as structures that were larger than 1 mm2 within or adjacent to the vessel lumen. The plaques were clearly distinguished from the lumen and surrounding epicardial fat. The plaque type was classified as follows: (a) Plaques containing calcified tissue that comprised more than 50% of the plaque area (attenuation >130 HU on native images) were classified as calcified. (b) Plaques with less than 50% calcium in the plaque area were classified as mixed. (c) Plaques without any calcium were classified as noncalcified plaques [16]. 2.4. Statistical analysis
2.2. Clinical and laboratory evaluation In addition to a laboratory examination, each subject underwent a questionnaire assessment and an anthropometric assessment. Systolic and diastolic blood pressures were measured twice on the same day, and the mean of the two values was used in the analysis. Height and body weight were measured using a digital scale. Body mass index was calculated as weight divided by height (in meters) squared, and waist circumference was measured at the midpoint between the lower costal margin and the iliac crest. These values were measured by a well-trained nurse. Subjects were classified as current smokers if they had been smoking for at least 1 year. The laboratory evaluation included measurements for aspartate aminotransferase (AST), alanine aminotransferase (ALT), gammaglutamyl transpeptidase (GGT), total bilirubin, total cholesterol, triglyceride, high-density lipoprotein (HDL) cholesterol, lowdensity lipoprotein (LDL) cholesterol, fasting glucose, high sensitive c-reactive protein (hs-CRP), hepatitis B surface antigen, and an
The data were expressed as the mean standard deviation or as a median with an interquartile range for continuous variables. Comparisons between groups were performed using the Chisquared test for categorical variables, and continuous data were analyzed using Student’s t-test or the ManneWhitney U test and the analysis of variance or the KruskaleWallis test, as was appropriate. To explore the association between bilirubin and coronary stenosis in a dose-dependent manner, serum bilirubin was stratified into quartiles (1st quartile: 0.8 mg/dL, 2nd quartile: 0.8e 1.0 mg/dL, 3rd quartile: 1.0e1.2 mg/dL, 4th quartile: >1.2 mg/dL). Logistic regression analysis was used to analyze the association between serum bilirubin and both coronary stenosis and each type of plaque, while also controlling for potential confounders. Covariates in the multivariable model, which were chosen for clinical importance as well as statistical significance, included age, sex, body mass index, waist circumference, diabetes, hypertension, total cholesterol, triglycerides, HDL cholesterol, daily alcohol
244
S.J. Kang et al. / Atherosclerosis 230 (2013) 242e248
consumption, and smoking. The independent relationship between bilirubin and the number of stenosis and plaque was examined by multiple regression analysis and determination of the standardized correlation coefficients. Statistical significance was defined as p < 0.05. All statistical analyses were performed using SPSS version 19.0 (SPSS Inc., Chicago, IL).
3. Results A total of 2862 male subjects met the inclusion criteria for this study. The majority of these subjects (n ¼ 2122, 74.1%) had no demonstrable atherosclerosis or stenosis in the coronary arteries, whereas the remaining 740 subjects (25.9%) had coronary atherosclerosis. Among them, 458 patients (61.9%) had insignificant stenosis and 282 patients (38.1%) showed significant coronary stenosis. The characteristics of the study population are presented in Table 1. Some of the more noticeable differences were observed in the mean age, prevalence of diabetes, hypertension, levels of bilirubin, HDL cholesterol, fasting glucose, and alcohol consumption between no stenosis and coronary atherosclerosis. Relationships between the prevalence of coronary atherosclerosis and stenosis according to serum bilirubin quartiles were illustrated in Supplementary Fig. 1. The proportion of patients with coronary atherosclerosis progressively decreased from 29.3% in subjects in the 1st quartile of serum bilirubin to 19.9% for those in the 4th quartile of serum bilirubin (p < 0.001). Similar trends were
observed in subjects with stenosis <50% and with stenosis 50% (p < 0.001 and p ¼ 0.01, respectively). To identify whether higher than normal levels of serum bilirubin were associated with coronary stenosis and plaques, serum total bilirubin concentration was classified into the elevated bilirubin group (>1.2 mg/dL) vs. the normal bilirubin group (1.2 mg/dL). The characteristics of the individuals in each group are summarized in Table 2. No differences were observed in levels of LDL cholesterol, fasting glucose, body mass index and prevalence of hypertension in two groups. However, compared with normal bilirubin group, elevated bilirubin group had significantly lower levels of triglycerides, hs-CRP, waist circumference and lower prevalence of diabetes mellitus and dyslipidemia. Thus, we explored robustness of our findings by conducting a univariate and multivariable analysis for subjects without diabetes and for subjects without diabetes and dyslipidemia. These sensitivity analyses gave similar results to the primary analysis, with a significant inverse association between calcified plaques and bilirubin level (Supplementary Tables S1 and S2). As shown in Table 3, number of stenosis was significantly different between the two groups. The prevalence of coronary plaques was also significantly lower in the elevated serum bilirubin group. For each type of plaque, calcified and mixed plaques were less prevalent in the elevated serum bilirubin group; however, the prevalence of noncalcified plaques was not significantly different between the two groups. In Supplementary Table S3, a multivariable linear regression analysis between bilirubin level and number
Table 1 Comparison of baseline characteristics according to presence and severity of stenosis.
Age (years) Diabetes mellitus, N (%) Diabetes medication, N (%) Hypertension, N (%) HT medication, N (%) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Lipid-lowering drugs, N (%) Body mass index (kg/m2) Waist circumference (cm) Bilirubin AST (IU/L) ALT (IU/L) GGT (IU/L) Cholesterol (mg/dL) Triglycerides (mg/dL) HDL cholesterol (mg/dL) LDL cholesterol (mg/dL) Creatinine (mg/dL) Fasting glucose (mg/dL) C-reactive protein (mg/L) Alcohol consumption (/week) None, N (%) 0< and 150 g, N (%) >150 g, N (%) Smoking Non-current smoker, N (%) Current smoker, N (%) CAC FRS 10 Year CAD risk 10%, N (%)
No stenosis (n ¼ 2122)
Coronary atherosclerosis (n ¼ 740)
Coronary atherosclerosis Stenosis < 50% (n ¼ 458)
Stenosis 50% (n ¼ 282)
54.0 8.1 283 (13.3) 173 (8.2) 732 (34.5) 607 (28.6) 120.5 13.4 80.2 10.2 517 (24.4) 25.0 2.6 89.0 6.8 1.0 (0.8, 1.3) 23.0 (20.0, 28.0) 25.0 (18.3, 35.0) 35.0 (24.0, 56.0) 195.9 34.2 121.0 (86.0, 172.0) 50.3 11.9 122.6 31.6 1.11 0.12 100.0 (93.0, 110.0) 0.06 (0.01, 0.16)
59.6 8.3b 173 (23.4)b 134 (18.1)b 392 (53.0)b 343 (46.4)b 124.5 14.6b 81.1 11.0 224 (30.3)b 25.1 2.5 89.7 6.9a 1.0 (0.8, 1.2)b 24.0 (20.0, 29.0) 25.0 (18.5, 35.0) 33.0 (23.0, 54.0) 193.4 34.6 128.0 (88.3, 179.0) 49.0 11.0b 120.4 32.7 1.11 0.14 103.0 (94.0, 118.3)b 0.07 (0.01, 0.17)
59.2 8.2 100 (21.8) 75 (16.4) 229 (50.0) 200 (43.7) 123.6 14.6 81.1 11.1 141 (30.8) 25.1 2.6 89.7 6.9 1.0 (0.8, 1.2) 23.0 (20.0, 29.0) 25.0 (19.0, 35.0) 33.0 (23.0, 55.0) 193.9 34.1 125.0 (87.0, 176.0) 49.9 11.3 120.6 31.5 1.10 0.13 102.5 (93.0, 117.3) 0.06 (0.01, 0.16)
60.1 8.5 73 (25.9) 59 (20.9) 163 (57.8)c 143 (50.7) 126.0 14.5c 81.1 10.8 83 (29.4) 25.1 2.5 89.7 6.9 1.0 (0.8, 1.2) 24.0 (19.0, 29.0) 25.0 (18.0, 36.0) 32.0 (23.5, 51.5) 192.5 35.3 129.0 (91.0, 185.0) 47.4 10.4d 120.0 34.7 1.12 0.15 105.0 (95.0, 121.0)c 0.08 (0.01, 0.19)
429 (20.2) 1112 (52.4) 581 (27.4)
191 (25.8)b 398 (53.8) 151 (20.4)
104 (22.7) 252 (55.0) 102 (22.3)
87 (30.9)c 146 (51.8) 49 (17.4)
1523 (71.8) 599 (28.2) 0.00 (0.00, 9.45) 4.6 3.2 93 (4.4)
566 (76.5)a 174 (23.5) 118.05 (30.43, 337.68)b 5.9 3.2b 82(11.1)b
350 (76.4) 108 (23.6) 82.75 (23.43, 229.7) 5.7 3.1 44 (9.6)
216 (76.6) 66 (23.4) 227.95 (63.70, 597.30)d 6.1 3.3 38 (13.5)
AST, aspartate aminotransferase; ALT, alanine aminotransferase; GGT, gamma-glutamyl transferase; HDL, high-density lipoprotein; LDL, low-density lipoprotein; CAC, coronary artery calcification; FRS, Framingham risk score; CAD, coronary artery disease; Data are shown as the mean SD or median (interquartile range). a p < 0.05 for no stenosis vs. coronary atherosclerosis. b p < 0.01 for no stenosis vs. coronary atherosclerosis. c p < 0.05 for stenosis < 50% vs. stenosis 50%. d p < 0.01 for stenosis < 50% vs. stenosis 50%.
S.J. Kang et al. / Atherosclerosis 230 (2013) 242e248
245
Table 2 Characteristics of study subjects, presence of stenosis and plaques according to the bilirubin level.
Age (years) Diabetes mellitus, N (%) Diabetes medication, N (%) Hypertension, N (%) HT medication, N (%) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Lipid-lowering drugs, N (%) Body mass index (kg/m2) Waist circumference (cm) Bilirubin (mg/dL) AST (IU/L) ALT (IU/L) GGT (IU/L) Cholesterol (mg/dL) Triglycerides (mg/dL) HDL cholesterol (mg/dL) LDL cholesterol (mg/dL) Creatinine (mg/dL) Fasting glucose (mg/dL) C-reactive protein (mg/L) Alcohol consumption (/week) None, N (%) 0< and 150 g, N (%) >150 g, N (%) Smoking Non-current smoke, N (%) Current smoker, N (%) CAC FRS 10 Year CAD risk 10%, N (%)
Total (n ¼ 2862)
Bilirubin 1.2 (n ¼ 2119)
Bilirubin > 1.2 (n ¼ 743)
p-Value
55.4 8.5 453 (15.8) 304 (10.6) 1119 (39.1) 945 (33.0) 121.5 13.8 80.4 10.4 741 (25.9) 25.0 2.6 89.2 6.8 1.0 (0.8, 1.3) 23.0 (20.0, 29.0) 25.0 (18.5, 35.0) 34.0 (24.0, 55.0) 195.2 34.3 123.0 (87.0, 175.0) 49.9 11.7 122.0 31.9 1.11 0.13 106.2 22.5 0.07 (0.01, 0.16)
55.9 8.5 364 (17.2) 255 (12.0) 829 (39.1) 718 (33.9) 121.3 13.9 80.0 10.4 571 (26.9) 25.1 2.6 89.4 6.9 0.9 (0.8, 1.1) 23.0 (19.0, 28.0) 24.0 (18.0, 35.0) 34.0 (24.0, 56.0) 194.5 34.2 124.0 (88.0, 178.0) 49.5 11.6 121.5 31.5 1.10 0.13 106.8 23.5 0.07 (0.01, 0.16)
54.1 8.2 89 (12.0) 49 (6.6) 290 (39.0) 227 (38.2) 122.2 13.5 81.6 10.4 170 (22.9) 25.0 2.6 88.6 6.8 1.5 (1.3, 1.7) 24.0 (20.0, 29.5) 26.0 (19.0, 35.0) 34.0 (24.0, 55.0) 197.4 34.5 118.0 (83.0, 164.0) 51.2 11.9 123.0 32.6 1.12 0.12 104.6 19.4 0.05 (0.01, 0.15)
<0.001 <0.001 <0.001 0.471 0.472 0.121 <0.001 0.030 0.527 0.013 <0.001a 0.002a 0.048a 0.868a 0.048 0.001a 0.001 0.255 0.001 0.171 0.005a 0.040
596 (20.8) 1455 (50.8) 705 (24.6)
464 (21.9) 1054 (49.7) 521 (24.6)
132 (17.8) 401 (54.0) 184 (24.8)
2029 (70.9) 750 (26.2) 0.00 (0.00, 55.40) 4.92 3.24 175 (6.2)
1466 (69.2) 595 (28.1) 0.00 (0.00, 64.20) 5.03 3.22 139 (6.7)
563 (75.8) 155 (20.9) 0.00 (0.00, 37.00) 4.60 3.26 36 (4.9)
<0.001
<0.001a 0.002 0.080a
AST, aspartate aminotransferase; ALT, alanine aminotransferase; GGT, gamma-glutamyl transferase; HDL, high-density lipoprotein; LDL, low-density lipoprotein; CAC, coronary artery calcification; FRS, Framingham risk score; CAD, coronary artery disease; Data are shown as the mean SD or median (interquartile range). a ManneWhitney U-test.
of stenosis and plaque showed that both number of stenosis and plaque were significantly associated with bilirubin level after adjustment for cardiovascular risk factors (p ¼ 0.046 and p ¼ 0.001, respectively). Multivariable logistic regression analysis, adjusted for known cardiovascular risk factors, showed that serum bilirubin levels were inversely associated with an increased prevalence of coronary atherosclerosis (odds ratio [OR] 0.70, 95% confidence interval [CI] 0.54e0.91, p ¼ 0.009, Table 4). In addition, multivariable analysis showed that coronary atherosclerosis was inversely associated with bilirubin grade in a dose-dependent manner (OR 0.73, 95% CI 0.56e 0.94 in 4th quartile vs. 1st quartile, p < 0.001 for trend). Similar relationships were observed between stenosis <50% and bilirubin. However, in the case of stenosis 50%, the relationship between the stenosis and the amount of serum bilirubin was attenuated and not statistically significant.
In Table 5, multivariable analysis showed that the presence of coronary plaque lesions was inversely associated with an increase of serum bilirubin after adjusting for risk factors (OR 0.72, 95% CI 0.57e0.90). Analysis for each type of plaques showed that the bilirubin level was inversely associated with the presence of calcified plaques after adjusting for risk factors (OR 0.66, 95% CI 0.52e0.84). Mixed and noncalcified plaques were not significantly associated with the serum bilirubin levels (OR 0.84, 95% CI 0.60e 1.19 and OR 1.00, 95% CI 0.69e1.46, respectively). 4. Discussion In this large and apparently healthy cohort, we investigated the relationships between serum bilirubin and both coronary artery stenosis and plaques. Elevated serum bilirubin was inversely associated with both coronary stenosis and the presence of calcified
Table 3 Presence of stenosis and plaques and number of stenosis and plaques in subjects with elevated serum bilirubin (>1.2 mg/dL) vs. normal serum bilirubin (1.2 mg/dL).
Coronary atherosclerosis, N (%) Stenosis < 50%, N (%) Stenosis 50%, N (%) Number of stenosisa Any plaque, N (%) Calcified plaque, N (%) Mixed plaque, N (%) Noncalcified plaque, N (%) Number of plaquesa a
Data are shown as the mean standard deviation.
Total (n ¼ 2862)
Bilirubin 1.2 (n ¼ 2119)
Bilirubin > 1.2 (n ¼ 743)
p-Value
740 (25.9) 458 (16.0) 282 (9.9) 0.54 1.14 1181 (41.3) 901 (31.5) 334 (11.7) 251 (8.8) 0.87 1.32
592 (27.9) 373 (17.6) 219 (10.3) 0.59 1.17 929 (43.8) 710 (33.5) 268 (12.6) 197 (9.3) 0.95 1.37
148 (19.9) 85 (11.4) 63 (8.5) 0.44 1.06 252 (33.9) 191 (25.7) 66 (8.9) 54 (7.3) 0.73 1.22
<0.001 <0.001 0.044 0.001 <0.001 <0.001 0.006 0.095 <0.001
246
S.J. Kang et al. / Atherosclerosis 230 (2013) 242e248
Table 4 Univariate and multivariable binary and ordinal analyses of the risk for atherosclerosis according to serum bilirubin level. Univariate model
Coronary atherosclerosis Bilirubina Bilirubin quartile Quartile 1 (0.8 mg/dL) Quartile 2 (0.9e1.0 mg/dL) Quartile 3 (1.1e1.2 mg/dL) Quartile 4 (>1.2 mg/dL) Bilirubin group Normal (1.2 mg/dL) Elevated (>1.2 mg/dL) Stenosis < 50% Bilirubina Bilirubin quartile Quartile 1 (0.8 mg/dL) Quartile 2 (0.9e1.0 mg/dL) Quartile 3 (1.1e1.2 mg/dL) Quartile 4 (>1.2 mg/dL) Bilirubin group Normal (1.2 mg/dL) Elevated (>1.2 mg/dL) Stenosis 50% Bilirubina Bilirubin quartile Quartile 1 (0.8 mg/dL) Quartile 2 (0.9e1.0 mg/dL) Quartile 3 (1.1e1.2 mg/dL) Quartile 4 (>1.2 mg/dL) Bilirubin group Normal (1.2 mg/dL) Elevated (>1.2 mg/dL)
Multivariable model
Odds ratio
95% Confidence interval
p Value
Odds ratio
95% Confidence interval
0.58
0.46e0.74
<0.001
0.70
0.54e0.91
0.009
1 0.95 0.94 0.73
0.75e1.22 0.73e1.23 0.56e0.94
<0.001b 0.701 0.692 0.015
b
p Value
1 0.93 0.84 0.62
0.75e1.16 0.66e1.07 0.49e0.78
<0.001 0.533 0.162 <0.001
1 0.64
0.52e0.79
<0.001
1 0.73
0.58e0.91
0.005
0.52
0.39e0.70
<0.001
0.62
0.45e0.86
0.004
1 0.97 0.86 0.57
0.75e1.26 0.65e1.15 0.42e0.76
<0.001b 0.815 0.310 <0.001
1 0.99 0.98 0.65
0.75e1.32 0.72e1.33 0.47e0.89
0.003b 0.960 0.883 0.007
1 0.59
0.45e0.75
<0.001
1 0.64
0.49e0.85
0.002
0.69
0.50e0.98
0.037
0.91
0.63e1.33
0.636
1 0.94 0.98 0.91
0.66e1.35 0.66e1.46 0.63e1.32
0.685b 0.747 0.932 0.625
1 0.90
0.65e1.24
0.507
1 0.87 0.81 0.71
0.63e1.21 0.57e1.16 0.51e0.99
0.032 0.416 0.254 0.044
1 0.74
0.55e0.99
0.044
b
Multivariate model was adjusted for age, body mass index, waist circumference, alcohol consumption, smoking status, diabetes, hypertension, total cholesterol, triglyceride and HDL cholesterol. a Analyzed as continuous variable. b p Value for the test of trend of odds.
plaques after adjusting for the traditional cardiovascular risk factors. Additionally, bilirubin levels were inversely associated with extent and severity of coronary atherosclerosis in a dosedependent manner, regardless of known metabolic risk factors. An increasing number of studies have suggested that high serum bilirubin is an independent protective factor for systemic atherosclerotic burdens such as carotid plaque [18], carotid intimal thickness [7], and peripheral artery disease [8]. A recent study reported that the CAC score is a better predictor of subsequent cardiovascular events than carotid intimal thickness [19]. Currently, two published papers have addressed the association of serum bilirubin with CAC. Tanaka et al. reported a significant inverse relationship between serum bilirubin and CAC in Japan; however the study population consisted of patients who had symptoms or clinical signs of coronary artery disease at the time of diagnosis [11]. Zhang et al. also suggested that total serum bilirubin was inversely related with CAC in Korean men [12]. Although the CAC score can provide an estimate of total coronary atherosclerosis, the relation of CAC to angiographic stenosis and the probability of vulnerable plaque or plaque rupture are unknown. Furthermore, the correlation between CAC and acute coronary events may be suboptimal [20]. Coronary CT was found to be more accurate than the calcium score in demonstrating coronary stenosis in coronary vessels and segments, especially in patients with a calcium score less than 400 [21]. Furthermore, coronary CT has been found to detect coronary plaques and more reliably differentiate the composition of plaques, compared with intravascular ultrasound [22]. Noncalcified plaques detected by multi-detector CT were found to be the strongest predictor of cardiac events, regardless of severity, and were found to act as a potential marker of plaque vulnerability [23]. To the best of our knowledge, the present study is the first report on the
association of total bilirubin with coronary artery stenosis, and various coronary plaques in the large screening population. A possible mechanism that could explain the inverse relationship between bilirubin and coronary stenosis is antioxidant effect of bilirubin. The oxidation of LDL cholesterol has a detrimental role in the pathogenesis of atherosclerosis [24]. The uptake of oxidized LDL cholesterol by macrophages results in the formation of oxygen and peroxyl radicals. Those radicals cause the atherosclerotic process and inflammation. Antioxidants may have a protective role from atherosclerosis by preventing oxidative modification of LDL cholesterol. In several studies, all forms of bilirubin were found to be effective in protecting LDL cholesterol against peroxidation [5,25]. Oxidized LDL cholesterol induces vascular cell gene expression to promote endogenous antioxidant defenses. Bilirubin and biliverdin, both powerful chain-breaking antioxidants, are metabolites of heme oxygenases pathway which is induced by oxidative stress and have anti-atherogenic effect [26]. In addition to antioxidant effect, an involvement of bilirubin in inflammatory reaction could be the other possible mechanism for bilirubin action in preventing atherosclerosis. Of the numerous circulating markers studied, hs-CRP is known as one of the most consistent marker of the risk of cardiovascular diseases [27]. We also showed significant difference of hs-CRP levels between the higher and lower bilirubin groups (Table 2). These findings imply that bilirubin, by preventing inflammation, might affect hs-CRP levels. In addition to evaluating coronary artery stenosis, coronary artery plaques, which are thought to be the direct causes of coronary artery disease, were also evaluated in this study. Calcified, mixed, and noncalcified plaques were diagnosed in 31.5%, 11.7%, and 8.8% of the subjects, respectively. As this study included elderly males, this proportion of plaque composition was consistent with previous
S.J. Kang et al. / Atherosclerosis 230 (2013) 242e248
247
Table 5 Univariate and multivariable binary and ordinal analyses of the risk for coronary plaques according to serum bilirubin level. Univariate model
Any plaques Bilirubina Bilirubin quartile Quartile 1 (0.8 mg/dL) Quartile 2 (0.9e1.0 mg/dL) Quartile 3 (1.1e1.2 mg/dL) Quartile 4 (>1.2 mg/dL) Bilirubin group Normal (1.2 mg/dL) Elevated (>1.2 mg/dL) Calcified plaques Bilirubina Bilirubin quartile Quartile 1 (0.8 mg/dL) Quartile 2 (0.9e1.0 mg/dL) Quartile 3 (1.1e1.2 mg/dL) Quartile 4 (>1.2 mg/dL) Bilirubin group 1.2 mg/dL >1.2 mg/dL Mixed plaques Bilirubina Bilirubin quartile Quartile 1 (0.8 mg/dL) Quartile 2 (0.9e1.0 mg/dL) Quartile 3 (1.1e1.2 mg/dL) Quartile 4 (>1.2 mg/dL) Bilirubin group 1.2 mg/dL >1.2 mg/dL Noncalcified plaques Bilirubina Bilirubin quartile Quartile 1 (0.8 mg/dL) Quartile 2 (0.9e1.0 mg/dL) Quartile 3 (1.1e1.2 mg/dL) Quartile 4 (>1.2 mg/dL) Bilirubin group 1.2 mg/dL >1.2 mg/dL
Multivariable model
Odds ratio
95% Confidence interval
p Value
Odds ratio
95% Confidence interval
0.61
0.50e0.75
<0.001
0.72
0.57e0.90
0.004
b
p Value
1 0.83 0.86 0.59
0.68e1.01 0.69e1.07 0.48e0.73
<0.001 0.068 0.171 <0.001
1 0.82 0.93 0.69
0.66e1.02 0.74e1.18 0.55e0.87
0.006b 0.080 0.564 0.001
1 0.66
0.55e0.78
<0.001
1 0.75
0.62e0.91
0.004
0.58
0.47e0.73
<0.001
0.66
0.52e0.84
0.001
1 0.81 0.69 0.60
0.66e0.99 0.55e0.87 0.49e0.75
<0.001b 0.049 0.002 <0.001
1 0.80 0.73 0.66
0.64e1.00 0.57e0.93 0.53e0.84
<0.001b 0.050 0.011 0.001
1 0.69
0.57e0.83
<0.001
1 0.76
0.62e0.93
0.008
0.69
0.50e0.95
0.022
0.84
0.60e1.19
0.326
b
1 0.79 0.96 0.62
0.58e1.07 0.70e1.31 0.45e0.85
0.022 0.127 0.795 0.003
1 0.84 1.10 0.77
0.61e1.16 0.79e1.54 0.55e1.09
0.326b 0.283 0.569 0.137
1 0.68
0.51e0.90
0.006
1 0.80
0.59e1.07
0.133
0.88
0.62e1.25
0.484
1.00
0.69e1.46
0.998
1 1.00 1.08 0.78
0.71e1.41 0.75e1.55 0.54e1.12
0.484b 0.989 0.684 0.183
1 0.96 1.08 0.86
0.67e1.38 0.73e1.59 0.58e1.26
0.998b 0.814 0.699 0.439
1 0.77
0.56e1.05
0.096
1 0.85
0.61e1.19
0.353
Multivariable model was adjusted for age, body mass index, waist circumference, alcohol consumption, smoking status, diabetes, hypertension, total cholesterol, triglyceride, and HDL cholesterol. a Analyzed as continuous variable. b p Value for the test of trend of odds.
reports that stated that calcified plaque was the most common lesion, followed by mixed and noncalcified plaques, in men older than 55 [28]. In our study, after adjusting for risk factors, a higher serum bilirubin level was inversely associated with the presence of calcified plaque; however, a correlation between serum bilirubin and noncalcified plaque was not significant. This may have been due to having only a relatively small number of subjects with noncalcified plaque (n ¼ 251) compared with the number of subjects with calcified plaques (n ¼ 901), per our asymptomatic screening policy. Association between bilirubin and calcified plaque can explain the relationships of bilirubin and CAC which are presented in former studies [11,12]. Our study also had some limitations. First, the design of this study was cross-sectional, which made it difficult to determine any causal or temporal relationships between high serum bilirubin and either coronary artery stenosis or the development of coronary plaques. Second, in this study we used a 16-detector row CT for the detection of coronary lesions. However, a meta-analysis comparing the diagnostic performance of 64-detector CT, 16-detector CT, and 4-detector CT reported that the diagnostic performance of CT significantly improved with newer generations of multi-detector CT scanners (64- and 16-detector vs. 4-detector CT) but no
significant difference in diagnostic performance was found between 64-detector and 16-detector CT units [29]. Third, we did not have data on the level of fasting insulin and adipocytokines; therefore, we could not evaluate the influences of insulin resistance and adipocytokines on coronary lesions. Finally, because our cohort only included men, it is difficult to generalize our results for women. In conclusion, subjects with elevated levels of serum bilirubin were found to have a lower risk for coronary artery stenosis and the presence of coronary plaques regardless of classical cardiovascular risk factors, compared with subjects with normal serum bilirubin levels. However, longitudinal studies are required to determine the effect of these findings on future cardiovascular events. Disclosures There are no conflicts of interest to disclosure. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.atherosclerosis.2013.06.021.
248
S.J. Kang et al. / Atherosclerosis 230 (2013) 242e248
References [1] Ross R. The pathogenesis of atherosclerosis: a perspective for the 1990s. Nature 1993;362:801e9. [2] Madamanchi NR, Vendrov A, Runge MS. Oxidative stress and vascular disease. Arteriosclerosis, Thrombosis, and Vascular Biology 2005;25:29e38. [3] Ceaser EK, Ramachandran A, Levonen AL, Darley-Usmar VM. Oxidized lowdensity lipoprotein and 15-deoxy-delta 12,14-PGJ2 increase mitochondrial complex I activity in endothelial cells. American Journal of Physiology, Heart and Circulatory Physiology 2003;285:H2298e308. [4] 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;341:454e7. [5] Stocker R, Yamamoto Y, McDonagh AF, Glazer AN, Ames BN. Bilirubin is an antioxidant of possible physiological importance. Science 1987;235:1043e6. [6] 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. Arteriosclerosis, Thrombosis, and Vascular Biology 1996;16: 250e5. [7] 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 flowmediated vasodilation and increased carotid intima-media thickness in both men and women. Atherosclerosis 2006;184:431e7. [8] Perlstein TS, Pande RL, Beckman JA, Creager MA. Serum total bilirubin level and prevalent lower-extremity peripheral arterial disease: National Health and Nutrition Examination Survey (NHANES) 1999 to 2004. Arteriosclerosis, Thrombosis, and Vascular Biology 2008;28:166e72. [9] Lin JP, O’Donnell CJ, Schwaiger JP, Cupples LA, Lingenhel A, Hunt SC, et al. Association between the UGT1A1*28 allele, bilirubin levels, and coronary heart disease in the Framingham Heart Study. Circulation 2006;114:1476e81. [10] Chen YH, Chau LY, Chen JW, Lin SJ. Serum bilirubin and ferritin levels link heme oxygenase-1 gene promoter polymorphism and susceptibility to coronary artery disease in diabetic patients. Diabetes Care 2008;31:1615e20. [11] Tanaka M, Fukui M, Tomiyasu K, Akabame S, Nakano K, Hasegawa G, et al. Low serum bilirubin concentration is associated with coronary artery calcification (CAC). Atherosclerosis 2009;206:287e91. [12] Zhang ZY, Bian LQ, Kim SJ, Zhou CC, Choi YH. Inverse relation of total serum bilirubin to coronary artery calcification score detected by multidetector computed tomography in males. Clinical Cardiology 2012;35:301e6. [13] Lim S, Shin H, Lee Y, Yoon JW, Kang SM, Choi SH, et al. Effect of metabolic syndrome on coronary artery stenosis and plaque characteristics as assessed with 64-detector row cardiac CT. Radiology 2011;261:437e45. [14] Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 1997;20:1183e97. [15] Park HE, Kim MK, Choi SY, Lee W, Shin CS, Cho SH, et al. The prevalence and distribution of coronary artery calcium in asymptomatic Korean population. The International Journal of Cardiovascular Imaging 2012;28:1227e35. [16] Hausleiter J, Meyer T, Hadamitzky M, Kastrati A, Martinoff S, Schomig A. Prevalence of noncalcified coronary plaques by 64-slice computed
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24] [25]
[26]
[27]
[28]
[29]
tomography in patients with an intermediate risk for significant coronary artery disease. Journal of the American College of Cardiology 2006;48:312e8. Leber AW, Knez A, von Ziegler F, Becker A, Nikolaou K, Paul S, et al. Quantification of obstructive and nonobstructive coronary lesions by 64-slice computed tomography. Journal of the American College of Cardiology 2005 Jul 5;46(1):147e54. Ishizaka N, Ishizaka Y, Takahashi E, Yamakado M, Hashimoto H. High serum bilirubin level is inversely associated with the presence of carotid plaque. Stroke 2001;32:580e3. Folsom AR, Kronmal RA, Detrano RC, O’Leary DH, Bild DE, Bluemke DA, et al. Coronary artery calcification compared with carotid intima-media thickness in the prediction of cardiovascular disease incidence: the Multi-Ethnic Study of Atherosclerosis (MESA). Archives of Internal Medicine 2008;168:1333e9. Budoff MJ, Achenbach S, Blumenthal RS, Carr JJ, Goldin JG, Greenland P, et al. Assessment of coronary artery disease by cardiac computed tomography: a scientific statement from the American Heart Association Committee on Cardiovascular Imaging and Intervention, Council on Cardiovascular Radiology and Intervention, and Committee on Cardiac Imaging, Council on Clinical Cardiology. Circulation 2006;114:1761e91. Lau GT, Ridley LJ, Schieb MC, Brieger DB, Freedman SB, Wong LA, et al. Coronary artery stenoses: detection with calcium scoring, CT angiography, and both methods combined. Radiology 2005;235:415e22. Schroeder S, Kopp AF, Baumbach A, Meisner C, Kuettner A, Georg C, et al. Noninvasive detection and evaluation of atherosclerotic coronary plaques with multislice computed tomography. Journal of the American College of Cardiology 2001;37:1430e5. Russo V, Zavalloni A, Bacchi Reggiani ML, Buttazzi K, Gostoli V, Bartolini S, et al. Incremental prognostic value of coronary CT angiography in patients with suspected coronary artery disease. Circulation Cardiovascular imaging 2010;3:351e9. Aviram M. Modified forms of low density lipoprotein and atherosclerosis. Atherosclerosis 1993;98:1e9. Neuzil J, Strocker R. Free and albumin-bound bilirubin are efficient co-antioxidants for alpha-tocopherol, inhibiting plasma and low density lipoprotein lipid peroxidation. Journal of Biological Chemistry 1994;269:16712e9. Siow RC, Sato H, Mann GE. Heme oxygenase-carbon monoxide signaling pathway in atherosclerosis: anti-atherogenic actions of bilirubin and carbon monoxide? Cardiovascular Research 1999;41:385e94. Ridker PM, Cushman M, Stampfer MJ, Tracy R, Hennekens CH. Inflammation, aspirin, and the risk of cardiovascular disease in apparently healthy men. New England Journal of Medicine 1997;336:973e9. Ahmadi N, Nabavi V, Hajsadeghi F, Flores F, French WJ, Mao SS, et al. Mortality incidence of patients with non-obstructive coronary artery disease diagnosed by computed tomography angiography. The American Journal of Cardiology 2011;107:10e6. Vanhoenacker PK, Heijenbrok-Kal MH, Van Heste R, Decramer I, Van Hoe LR, Wijns W, et al. Diagnostic performance of multidetector CT angiography for assessment of coronary artery disease: meta-analysis. Radiology 2007;244: 419e28.
Contents lists available at ScienceDirect
Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis
Elevated serum bilirubin levels are inversely associated with coronary artery atherosclerosis Seung Joo Kang a, Donghee Kim a, *, Hyo Eun Park a, Goh Eun Chung a, Seung Ho Choi a, Su-Yeon Choi a, Whal Lee b, Joo Sung Kim a, Sang-Heon Cho a a
Department of Internal Medicine, Healthcare Research Institute, Seoul National University, Hospital Healthcare System Gangnam Center, Seoul, South Korea Department of Radiology, Seoul National University College of Medicine, Seoul, South Korea
b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 September 2012 Received in revised form 20 June 2013 Accepted 22 June 2013 Available online 20 July 2013
Background: Inverse correlations of high serum bilirubin with metabolic and cardiovascular disease have been suggested. However, anti-atherogenic effects of bilirubin have not been well-established in terms of the presence of plaques and stenosis identified in coronary computed tomography (CT). Methods: A cross-sectional study was conducted on 2862 men who were free of cardiovascular disease and underwent coronary CT as part of a routine medical screening examination. Coronary stenotic lesions were considered to be incidences of coronary atherosclerosis, and stenosis was classified as stenosis <50% or 50%, according to degree of stenosis. Results: The prevalences of coronary atherosclerosis and stenosis 50% in subjects with elevated bilirubin levels (>1.2 mg/dL) were lower than those in subjects with normal bilirubin levels (1.2 mg/dL) (19.9% vs. 27.9%, p < 0.001, 8.5% vs. 10.3%, p ¼ 0.044). Bilirubin was inversely associated with total plaques (odds ratio [OR] 0.59, 95% confidence interval [CI] 0.48e0.73 in the 4th quartile vs. 1st quartile) and calcified plaques (OR 0.60, 95% CI 0.49e0.75) in univariate analysis. After adjusting for traditional risk factors, it was found that coronary atherosclerosis (OR 0.73, 95% CI 0.56e0.94 in the 4th quartile vs. 1st quartile) and calcified plaque (OR 0.66, 95% CI 0.53e0.84) were inversely associated with the bilirubin grade in a dose-dependent manner. Conclusions: The serum bilirubin level was inversely associated with coronary atherosclerosis and calcified plaques in a dose-dependent manner. These results suggested that serum bilirubin could be used as a protective biomarker of coronary artery disease. Ó 2013 Elsevier Ireland Ltd. All rights reserved.
Keywords: Bilirubin Atherosclerosis Coronary stenosis Plaque Atherosclerotic
1. Introduction Cardiovascular disease is the most common cause of mortality in developed countries and accounts for up to one-third of all deaths worldwide [1]. The overproduction of oxygen free radicals from oxidative stress is known to mediate various signaling pathways that underlie vascular inflammation in atherogenesis, from the
Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; CAC, coronary artery calcification; CT, computed tomography; CI, confidence interval; GGT, gamma-glutamyl transpeptidase; HDL, high-density lipoprotein; hsCRP, high sensitive c-reactive protein; LDL, low-density lipoprotein; OR, odds ratio; TG, triglycerides. * Corresponding author. Department of Internal Medicine, Healthcare Research Institute, Seoul National University Hospital, Healthcare System Gangnam Center, 39FL., Gangnam Finance Center 737, Yeoksam-Dong, Gangnam-Gu, Seoul 135-984, South Korea. Tel.: þ82 02 2112 5574; fax: þ82 02 2112 5635. E-mail addresses: [email protected] (S.J. Kang), [email protected] (D. Kim). 0021-9150/$ e see front matter Ó 2013 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.atherosclerosis.2013.06.021
initiation of fatty streak development to ultimate plaque rupture [2]. Although the pathogenesis of atherosclerosis has not been thoroughly investigated, oxidative stress and DNA damage induced by oxidized low-density lipoprotein (LDL) cholesterol and by dietinduced hypercholesterolemia contribute to the progression of atherosclerosis [3]. Therefore, antioxidants are thought to serve a protective role against atherosclerosis and coronary artery disease by preventing the oxidative modification of LDL cholesterol [4]. Bilirubin is a potent antioxidant under physiological conditions and suppresses the oxidation of lipids and lipoproteins. Thus, bilirubin has been demonstrated in vitro to prevent plaque formation and subsequent formation of atherosclerosis [5]. A prospective cohort study suggested that low levels of bilirubin are correlated to premature coronary artery disease [6]. Previous studies have demonstrated the inverse relationship between serum total bilirubin concentration to peripheral artery disease and carotid intimaemedia thickness [7,8]. A genetic association between genes that influence serum bilirubin concentration such as heme
S.J. Kang et al. / Atherosclerosis 230 (2013) 242e248
oxygenase and uridine diphosphate-glucuronosyl transferase and coronary artery disease has been addressed in several studies [9,10]. In recent studies, a high serum bilirubin level was inversely associated with the presence of coronary artery calcification (CAC) [11,12]. However, CAC is limited in its prediction of noncalcified plaques, and as a result, it fails to represent the entire spectrum of atherosclerotic plaques. CAC screening is also inadequate in evaluating the degree of stenosis caused by plaque [13]. Although individuals with low bilirubin levels have a higher prevalence of coronary artery disease, there are few studies on the effect of serum total bilirubin on coronary artery stenosis and plaques in large populations. Thus, we investigated the effect of serum bilirubin on both coronary artery stenosis and coronary plaques, as assessed by coronary computed tomography (CT), in large, apparently healthy Korean males. 2. Methods 2.1. Study population and study design We performed a cross-sectional study in subjects who underwent a comprehensive medical check-up at Seoul National University Hospital, Healthcare System Gangnam Center from July 2006 to March 2010. Some of the subjects had voluntarily paid for a general health check-up, while others were supported through their employer. During this period, 3319 male subjects underwent a coronary CT examination for screening. All coronary CT and clinical and laboratory assessments were performed on the same day or within 6 months of each other. Out of the 3319 subjects, we excluded 146 subjects who had a history of heart attack, coronary artery disease (including acute myocardial infarction), angina, or congestive heart failure. We also excluded 249 subjects with at least one potential cause of chronic liver disease: 182 with positive hepatitis B surface antigen, 55 with positive hepatitis C antibody, and 12 with a history of other hepatitis or liver disease (e.g., primary biliary cirrhosis, autoimmune hepatitis, Wilson’s disease, etc.). An additional 34 subjects with abnormal biliary tracts (as observed in abdominal ultrasonography) and 28 subjects (mean standard deviation of creatinine: 1.61 0.26 mg/dL) with abnormal renal function (creatinine level > 1.4 mg/dL) were also excluded. Finally a total of 2862 subjects were enrolled in this study. The Institutional Review Board of Seoul National University Hospital approved the study protocol, and the study was conducted in accordance with the ethical guidelines of the 1975 Declaration of Helsinki.
243
antibody to the hepatitis C virus. Venous blood samples were taken from all subjects before 10 AM, following a 12-h overnight fast. Hypertension was defined as having a systolic blood pressure 140 mmHg, a diastolic blood pressure 90 mmHg and/or current use of antihypertensive medications. Subjects with a fasting plasma glucose level of 126 mg/dL or currently on anti-diabetic treatments were defined as having diabetes mellitus [14]. 2.3. Measurement of coronary artery stenosis and plaque by coronary artery CT Coronary CT was performed using a 16-row multi-slice CT (Sensation 16, Siemens Medical Systems, Erlangen, Germany) as previously described [15]. Briefly, after obtaining a topogram of the chest, a calcium score scan and angiography were performed using the retrospective method with a tube voltage of 120 kV and a 110 effective mAs tube current with a 200 mm field of view and ECGgated dose modulation. Prior to the CT scan, the patient’s heart rate was measured. Patients with a heart rate higher than 60 beats per minute were given a 50e100 mg dose of metoprolol. All of the images were analyzed by a radiologist who was blinded to the clinical and laboratory results of the subjects. Each lesion was identified using a multi-planar reconstruction technique and maximum intensity projection of the short-axis, two- and fourchamber views. The degree of stenosis and the plaque characterization were measured from the coronary CT images with a dedicated computer 3D workstation (Rapidia; Infinitt, Seoul, Korea). Coronary artery stenosis was estimated from the contrast enhanced portion of the coronary lumen. The stenosis was semi-automatically traced at the maximal stenotic site and then compared with the mean value for the proximal and distal reference sites [16]. Subjects with any coronary stenotic lesions were classified as having coronary atherosclerosis. Stenosis was classified as <50% or 50% based on the degree of stenosis. Number of stenosis was assessed using the 15-segment model of American Heart Association [17]. Plaques were identified as structures that were larger than 1 mm2 within or adjacent to the vessel lumen. The plaques were clearly distinguished from the lumen and surrounding epicardial fat. The plaque type was classified as follows: (a) Plaques containing calcified tissue that comprised more than 50% of the plaque area (attenuation >130 HU on native images) were classified as calcified. (b) Plaques with less than 50% calcium in the plaque area were classified as mixed. (c) Plaques without any calcium were classified as noncalcified plaques [16]. 2.4. Statistical analysis
2.2. Clinical and laboratory evaluation In addition to a laboratory examination, each subject underwent a questionnaire assessment and an anthropometric assessment. Systolic and diastolic blood pressures were measured twice on the same day, and the mean of the two values was used in the analysis. Height and body weight were measured using a digital scale. Body mass index was calculated as weight divided by height (in meters) squared, and waist circumference was measured at the midpoint between the lower costal margin and the iliac crest. These values were measured by a well-trained nurse. Subjects were classified as current smokers if they had been smoking for at least 1 year. The laboratory evaluation included measurements for aspartate aminotransferase (AST), alanine aminotransferase (ALT), gammaglutamyl transpeptidase (GGT), total bilirubin, total cholesterol, triglyceride, high-density lipoprotein (HDL) cholesterol, lowdensity lipoprotein (LDL) cholesterol, fasting glucose, high sensitive c-reactive protein (hs-CRP), hepatitis B surface antigen, and an
The data were expressed as the mean standard deviation or as a median with an interquartile range for continuous variables. Comparisons between groups were performed using the Chisquared test for categorical variables, and continuous data were analyzed using Student’s t-test or the ManneWhitney U test and the analysis of variance or the KruskaleWallis test, as was appropriate. To explore the association between bilirubin and coronary stenosis in a dose-dependent manner, serum bilirubin was stratified into quartiles (1st quartile: 0.8 mg/dL, 2nd quartile: 0.8e 1.0 mg/dL, 3rd quartile: 1.0e1.2 mg/dL, 4th quartile: >1.2 mg/dL). Logistic regression analysis was used to analyze the association between serum bilirubin and both coronary stenosis and each type of plaque, while also controlling for potential confounders. Covariates in the multivariable model, which were chosen for clinical importance as well as statistical significance, included age, sex, body mass index, waist circumference, diabetes, hypertension, total cholesterol, triglycerides, HDL cholesterol, daily alcohol
244
S.J. Kang et al. / Atherosclerosis 230 (2013) 242e248
consumption, and smoking. The independent relationship between bilirubin and the number of stenosis and plaque was examined by multiple regression analysis and determination of the standardized correlation coefficients. Statistical significance was defined as p < 0.05. All statistical analyses were performed using SPSS version 19.0 (SPSS Inc., Chicago, IL).
3. Results A total of 2862 male subjects met the inclusion criteria for this study. The majority of these subjects (n ¼ 2122, 74.1%) had no demonstrable atherosclerosis or stenosis in the coronary arteries, whereas the remaining 740 subjects (25.9%) had coronary atherosclerosis. Among them, 458 patients (61.9%) had insignificant stenosis and 282 patients (38.1%) showed significant coronary stenosis. The characteristics of the study population are presented in Table 1. Some of the more noticeable differences were observed in the mean age, prevalence of diabetes, hypertension, levels of bilirubin, HDL cholesterol, fasting glucose, and alcohol consumption between no stenosis and coronary atherosclerosis. Relationships between the prevalence of coronary atherosclerosis and stenosis according to serum bilirubin quartiles were illustrated in Supplementary Fig. 1. The proportion of patients with coronary atherosclerosis progressively decreased from 29.3% in subjects in the 1st quartile of serum bilirubin to 19.9% for those in the 4th quartile of serum bilirubin (p < 0.001). Similar trends were
observed in subjects with stenosis <50% and with stenosis 50% (p < 0.001 and p ¼ 0.01, respectively). To identify whether higher than normal levels of serum bilirubin were associated with coronary stenosis and plaques, serum total bilirubin concentration was classified into the elevated bilirubin group (>1.2 mg/dL) vs. the normal bilirubin group (1.2 mg/dL). The characteristics of the individuals in each group are summarized in Table 2. No differences were observed in levels of LDL cholesterol, fasting glucose, body mass index and prevalence of hypertension in two groups. However, compared with normal bilirubin group, elevated bilirubin group had significantly lower levels of triglycerides, hs-CRP, waist circumference and lower prevalence of diabetes mellitus and dyslipidemia. Thus, we explored robustness of our findings by conducting a univariate and multivariable analysis for subjects without diabetes and for subjects without diabetes and dyslipidemia. These sensitivity analyses gave similar results to the primary analysis, with a significant inverse association between calcified plaques and bilirubin level (Supplementary Tables S1 and S2). As shown in Table 3, number of stenosis was significantly different between the two groups. The prevalence of coronary plaques was also significantly lower in the elevated serum bilirubin group. For each type of plaque, calcified and mixed plaques were less prevalent in the elevated serum bilirubin group; however, the prevalence of noncalcified plaques was not significantly different between the two groups. In Supplementary Table S3, a multivariable linear regression analysis between bilirubin level and number
Table 1 Comparison of baseline characteristics according to presence and severity of stenosis.
Age (years) Diabetes mellitus, N (%) Diabetes medication, N (%) Hypertension, N (%) HT medication, N (%) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Lipid-lowering drugs, N (%) Body mass index (kg/m2) Waist circumference (cm) Bilirubin AST (IU/L) ALT (IU/L) GGT (IU/L) Cholesterol (mg/dL) Triglycerides (mg/dL) HDL cholesterol (mg/dL) LDL cholesterol (mg/dL) Creatinine (mg/dL) Fasting glucose (mg/dL) C-reactive protein (mg/L) Alcohol consumption (/week) None, N (%) 0< and 150 g, N (%) >150 g, N (%) Smoking Non-current smoker, N (%) Current smoker, N (%) CAC FRS 10 Year CAD risk 10%, N (%)
No stenosis (n ¼ 2122)
Coronary atherosclerosis (n ¼ 740)
Coronary atherosclerosis Stenosis < 50% (n ¼ 458)
Stenosis 50% (n ¼ 282)
54.0 8.1 283 (13.3) 173 (8.2) 732 (34.5) 607 (28.6) 120.5 13.4 80.2 10.2 517 (24.4) 25.0 2.6 89.0 6.8 1.0 (0.8, 1.3) 23.0 (20.0, 28.0) 25.0 (18.3, 35.0) 35.0 (24.0, 56.0) 195.9 34.2 121.0 (86.0, 172.0) 50.3 11.9 122.6 31.6 1.11 0.12 100.0 (93.0, 110.0) 0.06 (0.01, 0.16)
59.6 8.3b 173 (23.4)b 134 (18.1)b 392 (53.0)b 343 (46.4)b 124.5 14.6b 81.1 11.0 224 (30.3)b 25.1 2.5 89.7 6.9a 1.0 (0.8, 1.2)b 24.0 (20.0, 29.0) 25.0 (18.5, 35.0) 33.0 (23.0, 54.0) 193.4 34.6 128.0 (88.3, 179.0) 49.0 11.0b 120.4 32.7 1.11 0.14 103.0 (94.0, 118.3)b 0.07 (0.01, 0.17)
59.2 8.2 100 (21.8) 75 (16.4) 229 (50.0) 200 (43.7) 123.6 14.6 81.1 11.1 141 (30.8) 25.1 2.6 89.7 6.9 1.0 (0.8, 1.2) 23.0 (20.0, 29.0) 25.0 (19.0, 35.0) 33.0 (23.0, 55.0) 193.9 34.1 125.0 (87.0, 176.0) 49.9 11.3 120.6 31.5 1.10 0.13 102.5 (93.0, 117.3) 0.06 (0.01, 0.16)
60.1 8.5 73 (25.9) 59 (20.9) 163 (57.8)c 143 (50.7) 126.0 14.5c 81.1 10.8 83 (29.4) 25.1 2.5 89.7 6.9 1.0 (0.8, 1.2) 24.0 (19.0, 29.0) 25.0 (18.0, 36.0) 32.0 (23.5, 51.5) 192.5 35.3 129.0 (91.0, 185.0) 47.4 10.4d 120.0 34.7 1.12 0.15 105.0 (95.0, 121.0)c 0.08 (0.01, 0.19)
429 (20.2) 1112 (52.4) 581 (27.4)
191 (25.8)b 398 (53.8) 151 (20.4)
104 (22.7) 252 (55.0) 102 (22.3)
87 (30.9)c 146 (51.8) 49 (17.4)
1523 (71.8) 599 (28.2) 0.00 (0.00, 9.45) 4.6 3.2 93 (4.4)
566 (76.5)a 174 (23.5) 118.05 (30.43, 337.68)b 5.9 3.2b 82(11.1)b
350 (76.4) 108 (23.6) 82.75 (23.43, 229.7) 5.7 3.1 44 (9.6)
216 (76.6) 66 (23.4) 227.95 (63.70, 597.30)d 6.1 3.3 38 (13.5)
AST, aspartate aminotransferase; ALT, alanine aminotransferase; GGT, gamma-glutamyl transferase; HDL, high-density lipoprotein; LDL, low-density lipoprotein; CAC, coronary artery calcification; FRS, Framingham risk score; CAD, coronary artery disease; Data are shown as the mean SD or median (interquartile range). a p < 0.05 for no stenosis vs. coronary atherosclerosis. b p < 0.01 for no stenosis vs. coronary atherosclerosis. c p < 0.05 for stenosis < 50% vs. stenosis 50%. d p < 0.01 for stenosis < 50% vs. stenosis 50%.
S.J. Kang et al. / Atherosclerosis 230 (2013) 242e248
245
Table 2 Characteristics of study subjects, presence of stenosis and plaques according to the bilirubin level.
Age (years) Diabetes mellitus, N (%) Diabetes medication, N (%) Hypertension, N (%) HT medication, N (%) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Lipid-lowering drugs, N (%) Body mass index (kg/m2) Waist circumference (cm) Bilirubin (mg/dL) AST (IU/L) ALT (IU/L) GGT (IU/L) Cholesterol (mg/dL) Triglycerides (mg/dL) HDL cholesterol (mg/dL) LDL cholesterol (mg/dL) Creatinine (mg/dL) Fasting glucose (mg/dL) C-reactive protein (mg/L) Alcohol consumption (/week) None, N (%) 0< and 150 g, N (%) >150 g, N (%) Smoking Non-current smoke, N (%) Current smoker, N (%) CAC FRS 10 Year CAD risk 10%, N (%)
Total (n ¼ 2862)
Bilirubin 1.2 (n ¼ 2119)
Bilirubin > 1.2 (n ¼ 743)
p-Value
55.4 8.5 453 (15.8) 304 (10.6) 1119 (39.1) 945 (33.0) 121.5 13.8 80.4 10.4 741 (25.9) 25.0 2.6 89.2 6.8 1.0 (0.8, 1.3) 23.0 (20.0, 29.0) 25.0 (18.5, 35.0) 34.0 (24.0, 55.0) 195.2 34.3 123.0 (87.0, 175.0) 49.9 11.7 122.0 31.9 1.11 0.13 106.2 22.5 0.07 (0.01, 0.16)
55.9 8.5 364 (17.2) 255 (12.0) 829 (39.1) 718 (33.9) 121.3 13.9 80.0 10.4 571 (26.9) 25.1 2.6 89.4 6.9 0.9 (0.8, 1.1) 23.0 (19.0, 28.0) 24.0 (18.0, 35.0) 34.0 (24.0, 56.0) 194.5 34.2 124.0 (88.0, 178.0) 49.5 11.6 121.5 31.5 1.10 0.13 106.8 23.5 0.07 (0.01, 0.16)
54.1 8.2 89 (12.0) 49 (6.6) 290 (39.0) 227 (38.2) 122.2 13.5 81.6 10.4 170 (22.9) 25.0 2.6 88.6 6.8 1.5 (1.3, 1.7) 24.0 (20.0, 29.5) 26.0 (19.0, 35.0) 34.0 (24.0, 55.0) 197.4 34.5 118.0 (83.0, 164.0) 51.2 11.9 123.0 32.6 1.12 0.12 104.6 19.4 0.05 (0.01, 0.15)
<0.001 <0.001 <0.001 0.471 0.472 0.121 <0.001 0.030 0.527 0.013 <0.001a 0.002a 0.048a 0.868a 0.048 0.001a 0.001 0.255 0.001 0.171 0.005a 0.040
596 (20.8) 1455 (50.8) 705 (24.6)
464 (21.9) 1054 (49.7) 521 (24.6)
132 (17.8) 401 (54.0) 184 (24.8)
2029 (70.9) 750 (26.2) 0.00 (0.00, 55.40) 4.92 3.24 175 (6.2)
1466 (69.2) 595 (28.1) 0.00 (0.00, 64.20) 5.03 3.22 139 (6.7)
563 (75.8) 155 (20.9) 0.00 (0.00, 37.00) 4.60 3.26 36 (4.9)
<0.001
<0.001a 0.002 0.080a
AST, aspartate aminotransferase; ALT, alanine aminotransferase; GGT, gamma-glutamyl transferase; HDL, high-density lipoprotein; LDL, low-density lipoprotein; CAC, coronary artery calcification; FRS, Framingham risk score; CAD, coronary artery disease; Data are shown as the mean SD or median (interquartile range). a ManneWhitney U-test.
of stenosis and plaque showed that both number of stenosis and plaque were significantly associated with bilirubin level after adjustment for cardiovascular risk factors (p ¼ 0.046 and p ¼ 0.001, respectively). Multivariable logistic regression analysis, adjusted for known cardiovascular risk factors, showed that serum bilirubin levels were inversely associated with an increased prevalence of coronary atherosclerosis (odds ratio [OR] 0.70, 95% confidence interval [CI] 0.54e0.91, p ¼ 0.009, Table 4). In addition, multivariable analysis showed that coronary atherosclerosis was inversely associated with bilirubin grade in a dose-dependent manner (OR 0.73, 95% CI 0.56e 0.94 in 4th quartile vs. 1st quartile, p < 0.001 for trend). Similar relationships were observed between stenosis <50% and bilirubin. However, in the case of stenosis 50%, the relationship between the stenosis and the amount of serum bilirubin was attenuated and not statistically significant.
In Table 5, multivariable analysis showed that the presence of coronary plaque lesions was inversely associated with an increase of serum bilirubin after adjusting for risk factors (OR 0.72, 95% CI 0.57e0.90). Analysis for each type of plaques showed that the bilirubin level was inversely associated with the presence of calcified plaques after adjusting for risk factors (OR 0.66, 95% CI 0.52e0.84). Mixed and noncalcified plaques were not significantly associated with the serum bilirubin levels (OR 0.84, 95% CI 0.60e 1.19 and OR 1.00, 95% CI 0.69e1.46, respectively). 4. Discussion In this large and apparently healthy cohort, we investigated the relationships between serum bilirubin and both coronary artery stenosis and plaques. Elevated serum bilirubin was inversely associated with both coronary stenosis and the presence of calcified
Table 3 Presence of stenosis and plaques and number of stenosis and plaques in subjects with elevated serum bilirubin (>1.2 mg/dL) vs. normal serum bilirubin (1.2 mg/dL).
Coronary atherosclerosis, N (%) Stenosis < 50%, N (%) Stenosis 50%, N (%) Number of stenosisa Any plaque, N (%) Calcified plaque, N (%) Mixed plaque, N (%) Noncalcified plaque, N (%) Number of plaquesa a
Data are shown as the mean standard deviation.
Total (n ¼ 2862)
Bilirubin 1.2 (n ¼ 2119)
Bilirubin > 1.2 (n ¼ 743)
p-Value
740 (25.9) 458 (16.0) 282 (9.9) 0.54 1.14 1181 (41.3) 901 (31.5) 334 (11.7) 251 (8.8) 0.87 1.32
592 (27.9) 373 (17.6) 219 (10.3) 0.59 1.17 929 (43.8) 710 (33.5) 268 (12.6) 197 (9.3) 0.95 1.37
148 (19.9) 85 (11.4) 63 (8.5) 0.44 1.06 252 (33.9) 191 (25.7) 66 (8.9) 54 (7.3) 0.73 1.22
<0.001 <0.001 0.044 0.001 <0.001 <0.001 0.006 0.095 <0.001
246
S.J. Kang et al. / Atherosclerosis 230 (2013) 242e248
Table 4 Univariate and multivariable binary and ordinal analyses of the risk for atherosclerosis according to serum bilirubin level. Univariate model
Coronary atherosclerosis Bilirubina Bilirubin quartile Quartile 1 (0.8 mg/dL) Quartile 2 (0.9e1.0 mg/dL) Quartile 3 (1.1e1.2 mg/dL) Quartile 4 (>1.2 mg/dL) Bilirubin group Normal (1.2 mg/dL) Elevated (>1.2 mg/dL) Stenosis < 50% Bilirubina Bilirubin quartile Quartile 1 (0.8 mg/dL) Quartile 2 (0.9e1.0 mg/dL) Quartile 3 (1.1e1.2 mg/dL) Quartile 4 (>1.2 mg/dL) Bilirubin group Normal (1.2 mg/dL) Elevated (>1.2 mg/dL) Stenosis 50% Bilirubina Bilirubin quartile Quartile 1 (0.8 mg/dL) Quartile 2 (0.9e1.0 mg/dL) Quartile 3 (1.1e1.2 mg/dL) Quartile 4 (>1.2 mg/dL) Bilirubin group Normal (1.2 mg/dL) Elevated (>1.2 mg/dL)
Multivariable model
Odds ratio
95% Confidence interval
p Value
Odds ratio
95% Confidence interval
0.58
0.46e0.74
<0.001
0.70
0.54e0.91
0.009
1 0.95 0.94 0.73
0.75e1.22 0.73e1.23 0.56e0.94
<0.001b 0.701 0.692 0.015
b
p Value
1 0.93 0.84 0.62
0.75e1.16 0.66e1.07 0.49e0.78
<0.001 0.533 0.162 <0.001
1 0.64
0.52e0.79
<0.001
1 0.73
0.58e0.91
0.005
0.52
0.39e0.70
<0.001
0.62
0.45e0.86
0.004
1 0.97 0.86 0.57
0.75e1.26 0.65e1.15 0.42e0.76
<0.001b 0.815 0.310 <0.001
1 0.99 0.98 0.65
0.75e1.32 0.72e1.33 0.47e0.89
0.003b 0.960 0.883 0.007
1 0.59
0.45e0.75
<0.001
1 0.64
0.49e0.85
0.002
0.69
0.50e0.98
0.037
0.91
0.63e1.33
0.636
1 0.94 0.98 0.91
0.66e1.35 0.66e1.46 0.63e1.32
0.685b 0.747 0.932 0.625
1 0.90
0.65e1.24
0.507
1 0.87 0.81 0.71
0.63e1.21 0.57e1.16 0.51e0.99
0.032 0.416 0.254 0.044
1 0.74
0.55e0.99
0.044
b
Multivariate model was adjusted for age, body mass index, waist circumference, alcohol consumption, smoking status, diabetes, hypertension, total cholesterol, triglyceride and HDL cholesterol. a Analyzed as continuous variable. b p Value for the test of trend of odds.
plaques after adjusting for the traditional cardiovascular risk factors. Additionally, bilirubin levels were inversely associated with extent and severity of coronary atherosclerosis in a dosedependent manner, regardless of known metabolic risk factors. An increasing number of studies have suggested that high serum bilirubin is an independent protective factor for systemic atherosclerotic burdens such as carotid plaque [18], carotid intimal thickness [7], and peripheral artery disease [8]. A recent study reported that the CAC score is a better predictor of subsequent cardiovascular events than carotid intimal thickness [19]. Currently, two published papers have addressed the association of serum bilirubin with CAC. Tanaka et al. reported a significant inverse relationship between serum bilirubin and CAC in Japan; however the study population consisted of patients who had symptoms or clinical signs of coronary artery disease at the time of diagnosis [11]. Zhang et al. also suggested that total serum bilirubin was inversely related with CAC in Korean men [12]. Although the CAC score can provide an estimate of total coronary atherosclerosis, the relation of CAC to angiographic stenosis and the probability of vulnerable plaque or plaque rupture are unknown. Furthermore, the correlation between CAC and acute coronary events may be suboptimal [20]. Coronary CT was found to be more accurate than the calcium score in demonstrating coronary stenosis in coronary vessels and segments, especially in patients with a calcium score less than 400 [21]. Furthermore, coronary CT has been found to detect coronary plaques and more reliably differentiate the composition of plaques, compared with intravascular ultrasound [22]. Noncalcified plaques detected by multi-detector CT were found to be the strongest predictor of cardiac events, regardless of severity, and were found to act as a potential marker of plaque vulnerability [23]. To the best of our knowledge, the present study is the first report on the
association of total bilirubin with coronary artery stenosis, and various coronary plaques in the large screening population. A possible mechanism that could explain the inverse relationship between bilirubin and coronary stenosis is antioxidant effect of bilirubin. The oxidation of LDL cholesterol has a detrimental role in the pathogenesis of atherosclerosis [24]. The uptake of oxidized LDL cholesterol by macrophages results in the formation of oxygen and peroxyl radicals. Those radicals cause the atherosclerotic process and inflammation. Antioxidants may have a protective role from atherosclerosis by preventing oxidative modification of LDL cholesterol. In several studies, all forms of bilirubin were found to be effective in protecting LDL cholesterol against peroxidation [5,25]. Oxidized LDL cholesterol induces vascular cell gene expression to promote endogenous antioxidant defenses. Bilirubin and biliverdin, both powerful chain-breaking antioxidants, are metabolites of heme oxygenases pathway which is induced by oxidative stress and have anti-atherogenic effect [26]. In addition to antioxidant effect, an involvement of bilirubin in inflammatory reaction could be the other possible mechanism for bilirubin action in preventing atherosclerosis. Of the numerous circulating markers studied, hs-CRP is known as one of the most consistent marker of the risk of cardiovascular diseases [27]. We also showed significant difference of hs-CRP levels between the higher and lower bilirubin groups (Table 2). These findings imply that bilirubin, by preventing inflammation, might affect hs-CRP levels. In addition to evaluating coronary artery stenosis, coronary artery plaques, which are thought to be the direct causes of coronary artery disease, were also evaluated in this study. Calcified, mixed, and noncalcified plaques were diagnosed in 31.5%, 11.7%, and 8.8% of the subjects, respectively. As this study included elderly males, this proportion of plaque composition was consistent with previous
S.J. Kang et al. / Atherosclerosis 230 (2013) 242e248
247
Table 5 Univariate and multivariable binary and ordinal analyses of the risk for coronary plaques according to serum bilirubin level. Univariate model
Any plaques Bilirubina Bilirubin quartile Quartile 1 (0.8 mg/dL) Quartile 2 (0.9e1.0 mg/dL) Quartile 3 (1.1e1.2 mg/dL) Quartile 4 (>1.2 mg/dL) Bilirubin group Normal (1.2 mg/dL) Elevated (>1.2 mg/dL) Calcified plaques Bilirubina Bilirubin quartile Quartile 1 (0.8 mg/dL) Quartile 2 (0.9e1.0 mg/dL) Quartile 3 (1.1e1.2 mg/dL) Quartile 4 (>1.2 mg/dL) Bilirubin group 1.2 mg/dL >1.2 mg/dL Mixed plaques Bilirubina Bilirubin quartile Quartile 1 (0.8 mg/dL) Quartile 2 (0.9e1.0 mg/dL) Quartile 3 (1.1e1.2 mg/dL) Quartile 4 (>1.2 mg/dL) Bilirubin group 1.2 mg/dL >1.2 mg/dL Noncalcified plaques Bilirubina Bilirubin quartile Quartile 1 (0.8 mg/dL) Quartile 2 (0.9e1.0 mg/dL) Quartile 3 (1.1e1.2 mg/dL) Quartile 4 (>1.2 mg/dL) Bilirubin group 1.2 mg/dL >1.2 mg/dL
Multivariable model
Odds ratio
95% Confidence interval
p Value
Odds ratio
95% Confidence interval
0.61
0.50e0.75
<0.001
0.72
0.57e0.90
0.004
b
p Value
1 0.83 0.86 0.59
0.68e1.01 0.69e1.07 0.48e0.73
<0.001 0.068 0.171 <0.001
1 0.82 0.93 0.69
0.66e1.02 0.74e1.18 0.55e0.87
0.006b 0.080 0.564 0.001
1 0.66
0.55e0.78
<0.001
1 0.75
0.62e0.91
0.004
0.58
0.47e0.73
<0.001
0.66
0.52e0.84
0.001
1 0.81 0.69 0.60
0.66e0.99 0.55e0.87 0.49e0.75
<0.001b 0.049 0.002 <0.001
1 0.80 0.73 0.66
0.64e1.00 0.57e0.93 0.53e0.84
<0.001b 0.050 0.011 0.001
1 0.69
0.57e0.83
<0.001
1 0.76
0.62e0.93
0.008
0.69
0.50e0.95
0.022
0.84
0.60e1.19
0.326
b
1 0.79 0.96 0.62
0.58e1.07 0.70e1.31 0.45e0.85
0.022 0.127 0.795 0.003
1 0.84 1.10 0.77
0.61e1.16 0.79e1.54 0.55e1.09
0.326b 0.283 0.569 0.137
1 0.68
0.51e0.90
0.006
1 0.80
0.59e1.07
0.133
0.88
0.62e1.25
0.484
1.00
0.69e1.46
0.998
1 1.00 1.08 0.78
0.71e1.41 0.75e1.55 0.54e1.12
0.484b 0.989 0.684 0.183
1 0.96 1.08 0.86
0.67e1.38 0.73e1.59 0.58e1.26
0.998b 0.814 0.699 0.439
1 0.77
0.56e1.05
0.096
1 0.85
0.61e1.19
0.353
Multivariable model was adjusted for age, body mass index, waist circumference, alcohol consumption, smoking status, diabetes, hypertension, total cholesterol, triglyceride, and HDL cholesterol. a Analyzed as continuous variable. b p Value for the test of trend of odds.
reports that stated that calcified plaque was the most common lesion, followed by mixed and noncalcified plaques, in men older than 55 [28]. In our study, after adjusting for risk factors, a higher serum bilirubin level was inversely associated with the presence of calcified plaque; however, a correlation between serum bilirubin and noncalcified plaque was not significant. This may have been due to having only a relatively small number of subjects with noncalcified plaque (n ¼ 251) compared with the number of subjects with calcified plaques (n ¼ 901), per our asymptomatic screening policy. Association between bilirubin and calcified plaque can explain the relationships of bilirubin and CAC which are presented in former studies [11,12]. Our study also had some limitations. First, the design of this study was cross-sectional, which made it difficult to determine any causal or temporal relationships between high serum bilirubin and either coronary artery stenosis or the development of coronary plaques. Second, in this study we used a 16-detector row CT for the detection of coronary lesions. However, a meta-analysis comparing the diagnostic performance of 64-detector CT, 16-detector CT, and 4-detector CT reported that the diagnostic performance of CT significantly improved with newer generations of multi-detector CT scanners (64- and 16-detector vs. 4-detector CT) but no
significant difference in diagnostic performance was found between 64-detector and 16-detector CT units [29]. Third, we did not have data on the level of fasting insulin and adipocytokines; therefore, we could not evaluate the influences of insulin resistance and adipocytokines on coronary lesions. Finally, because our cohort only included men, it is difficult to generalize our results for women. In conclusion, subjects with elevated levels of serum bilirubin were found to have a lower risk for coronary artery stenosis and the presence of coronary plaques regardless of classical cardiovascular risk factors, compared with subjects with normal serum bilirubin levels. However, longitudinal studies are required to determine the effect of these findings on future cardiovascular events. Disclosures There are no conflicts of interest to disclosure. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.atherosclerosis.2013.06.021.
248
S.J. Kang et al. / Atherosclerosis 230 (2013) 242e248
References [1] Ross R. The pathogenesis of atherosclerosis: a perspective for the 1990s. Nature 1993;362:801e9. [2] Madamanchi NR, Vendrov A, Runge MS. Oxidative stress and vascular disease. Arteriosclerosis, Thrombosis, and Vascular Biology 2005;25:29e38. [3] Ceaser EK, Ramachandran A, Levonen AL, Darley-Usmar VM. Oxidized lowdensity lipoprotein and 15-deoxy-delta 12,14-PGJ2 increase mitochondrial complex I activity in endothelial cells. American Journal of Physiology, Heart and Circulatory Physiology 2003;285:H2298e308. [4] 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;341:454e7. [5] Stocker R, Yamamoto Y, McDonagh AF, Glazer AN, Ames BN. Bilirubin is an antioxidant of possible physiological importance. Science 1987;235:1043e6. [6] 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. Arteriosclerosis, Thrombosis, and Vascular Biology 1996;16: 250e5. [7] 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 flowmediated vasodilation and increased carotid intima-media thickness in both men and women. Atherosclerosis 2006;184:431e7. [8] Perlstein TS, Pande RL, Beckman JA, Creager MA. Serum total bilirubin level and prevalent lower-extremity peripheral arterial disease: National Health and Nutrition Examination Survey (NHANES) 1999 to 2004. Arteriosclerosis, Thrombosis, and Vascular Biology 2008;28:166e72. [9] Lin JP, O’Donnell CJ, Schwaiger JP, Cupples LA, Lingenhel A, Hunt SC, et al. Association between the UGT1A1*28 allele, bilirubin levels, and coronary heart disease in the Framingham Heart Study. Circulation 2006;114:1476e81. [10] Chen YH, Chau LY, Chen JW, Lin SJ. Serum bilirubin and ferritin levels link heme oxygenase-1 gene promoter polymorphism and susceptibility to coronary artery disease in diabetic patients. Diabetes Care 2008;31:1615e20. [11] Tanaka M, Fukui M, Tomiyasu K, Akabame S, Nakano K, Hasegawa G, et al. Low serum bilirubin concentration is associated with coronary artery calcification (CAC). Atherosclerosis 2009;206:287e91. [12] Zhang ZY, Bian LQ, Kim SJ, Zhou CC, Choi YH. Inverse relation of total serum bilirubin to coronary artery calcification score detected by multidetector computed tomography in males. Clinical Cardiology 2012;35:301e6. [13] Lim S, Shin H, Lee Y, Yoon JW, Kang SM, Choi SH, et al. Effect of metabolic syndrome on coronary artery stenosis and plaque characteristics as assessed with 64-detector row cardiac CT. Radiology 2011;261:437e45. [14] Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 1997;20:1183e97. [15] Park HE, Kim MK, Choi SY, Lee W, Shin CS, Cho SH, et al. The prevalence and distribution of coronary artery calcium in asymptomatic Korean population. The International Journal of Cardiovascular Imaging 2012;28:1227e35. [16] Hausleiter J, Meyer T, Hadamitzky M, Kastrati A, Martinoff S, Schomig A. Prevalence of noncalcified coronary plaques by 64-slice computed
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24] [25]
[26]
[27]
[28]
[29]
tomography in patients with an intermediate risk for significant coronary artery disease. Journal of the American College of Cardiology 2006;48:312e8. Leber AW, Knez A, von Ziegler F, Becker A, Nikolaou K, Paul S, et al. Quantification of obstructive and nonobstructive coronary lesions by 64-slice computed tomography. Journal of the American College of Cardiology 2005 Jul 5;46(1):147e54. Ishizaka N, Ishizaka Y, Takahashi E, Yamakado M, Hashimoto H. High serum bilirubin level is inversely associated with the presence of carotid plaque. Stroke 2001;32:580e3. Folsom AR, Kronmal RA, Detrano RC, O’Leary DH, Bild DE, Bluemke DA, et al. Coronary artery calcification compared with carotid intima-media thickness in the prediction of cardiovascular disease incidence: the Multi-Ethnic Study of Atherosclerosis (MESA). Archives of Internal Medicine 2008;168:1333e9. Budoff MJ, Achenbach S, Blumenthal RS, Carr JJ, Goldin JG, Greenland P, et al. Assessment of coronary artery disease by cardiac computed tomography: a scientific statement from the American Heart Association Committee on Cardiovascular Imaging and Intervention, Council on Cardiovascular Radiology and Intervention, and Committee on Cardiac Imaging, Council on Clinical Cardiology. Circulation 2006;114:1761e91. Lau GT, Ridley LJ, Schieb MC, Brieger DB, Freedman SB, Wong LA, et al. Coronary artery stenoses: detection with calcium scoring, CT angiography, and both methods combined. Radiology 2005;235:415e22. Schroeder S, Kopp AF, Baumbach A, Meisner C, Kuettner A, Georg C, et al. Noninvasive detection and evaluation of atherosclerotic coronary plaques with multislice computed tomography. Journal of the American College of Cardiology 2001;37:1430e5. Russo V, Zavalloni A, Bacchi Reggiani ML, Buttazzi K, Gostoli V, Bartolini S, et al. Incremental prognostic value of coronary CT angiography in patients with suspected coronary artery disease. Circulation Cardiovascular imaging 2010;3:351e9. Aviram M. Modified forms of low density lipoprotein and atherosclerosis. Atherosclerosis 1993;98:1e9. Neuzil J, Strocker R. Free and albumin-bound bilirubin are efficient co-antioxidants for alpha-tocopherol, inhibiting plasma and low density lipoprotein lipid peroxidation. Journal of Biological Chemistry 1994;269:16712e9. Siow RC, Sato H, Mann GE. Heme oxygenase-carbon monoxide signaling pathway in atherosclerosis: anti-atherogenic actions of bilirubin and carbon monoxide? Cardiovascular Research 1999;41:385e94. Ridker PM, Cushman M, Stampfer MJ, Tracy R, Hennekens CH. Inflammation, aspirin, and the risk of cardiovascular disease in apparently healthy men. New England Journal of Medicine 1997;336:973e9. Ahmadi N, Nabavi V, Hajsadeghi F, Flores F, French WJ, Mao SS, et al. Mortality incidence of patients with non-obstructive coronary artery disease diagnosed by computed tomography angiography. The American Journal of Cardiology 2011;107:10e6. Vanhoenacker PK, Heijenbrok-Kal MH, Van Heste R, Decramer I, Van Hoe LR, Wijns W, et al. Diagnostic performance of multidetector CT angiography for assessment of coronary artery disease: meta-analysis. Radiology 2007;244: 419e28.