Serum bilirubin distribution and its relation to cardiovascular risk in children and young adults

Atherosclerosis 131 (1997) 107 – 113

Serum bilirubin distribution and its relation to cardiovascular risk in children and young adults Malathy Madhavan a, Wendy A. Wattigney b, Sathanur R. Srinivasan b, Gerald S. Berenson b,* b

a Department of Pathology, Uni6ersity of Madras, Postgraduate Institute of Basic Medical Sciences, Madras 600 113, India Tulane Center for Cardio6ascular Health, Tulane Uni6ersity Medical Center, Tulane School of Public Health and Tropical Medicine, 1501 Canal Street, 14th Floor, New Orleans, LA 70 112 -2824, USA

Received 24 October 1996; received in revised form 17 January 1997; accepted 29 January 1997

Abstract There is evidence that bilirubin functions as an endogenous tissue protector by its antioxidant and anti-complement actions, properties that are relevant to atherogenesis. Serum bilirubin distribution and its relation to cardiovascular risk were examined in 4156 individuals aged 5–30 years from a biracial (black – white) community. Bilirubin levels showed significant differences related to race (whites \ blacks) and sex (males\ females, except in 5 – 10 year olds). In males the levels increased with age up to 24 years, while in females the changes were less conspicuous. Both adiposity and cigarette smoking associated independently and inversely with bilirubin. In addition, serum bilirubin correlated positively with HDL cholesterol and inversely with triglycerides, VLDL cholesterol, LDL cholesterol, insulin, glucose and systolic blood pressure although these correlations were significant only in certain age-race-sex groups. Offspring with a parental history of heart attack or hypertension had consistently lower bilirubin levels than those without such parental history. Thus, bilirubin may be an inverse risk factor for cardiovascular disease. © 1997 Elsevier Science Ireland Ltd. Keywords: Bilirubin; Cardiovascular risk factors; Children and young adults; Race (black – white)

1. Introduction It is well recognized that lipid peroxidation of lipoproteins is an important step in atherogenesis and that antioxidants are beneficial in attenuating this process [1]. Since bilirubin, the end product of heme degradation, is a potent scavenger of peroxyl radicals, it has been suggested that bilirubin functions as an antioxidant of physiologic significance [2 – 8]. In terms of atherogenesis, bilirubin could have a role in protecting lipids and lipoproteins against oxidation. Recently, Schwertner and co-workers [9] demonstrated an inverse association between levels of serum bilirubin and coronary artery disease (CAD), independent of other established risk factors. * Corresponding author. Tel.: +1 504 5857197; fax: + 1 504 5857194.

In view of the potential significance of bilirubin as an independent negative risk factor for CAD, there is a need to examine the distribution of serum bilirubin and its correlates in the general population. Further, the levels of serum bilirubin in asymptomatic young offspring may be related to cardiovascular disease in their parents, given the familial nature of risk factors [10– 12]. The present study examines these aspects in a biracial (black–white) population of children and young adults.

2. Methods

2.1. Population The Bogalusa Heart Study is a long-term epidemiologic study of the early natural history of cardiovascu-

0021-9150/97/$17.00 © 1997 Elsevier Science Ireland Ltd. All rights reserved. PII S 0 0 2 1 - 9 1 5 0 ( 9 7 ) 0 6 0 8 8 - 7

108

M. Madha6an et al. / Atherosclerosis 131 (1997) 107–113

lar disease in the biracial (65% white, 35% black) community of Bogalusa, LA. Serum bilirubin levels were measured as part of the post-high school surveys (1985–86, ages 18 – 26 years, and 1988 – 91, ages 20–31 years), the 1987 – 88 cross-sectional survey of school aged children, and the 1992 examination of high school seniors. Data were combined, including each individual once only at his/her most recent examination, to yield a sample size of 4930. Persons younger than 5 years (n =13) or older than 30 years (n = 45), race other than black or white (n = 5), one outlier with a bilirubin level of 9.4 mg/dl, and non-fasters (n =710) were excluded, leaving a total 4156 children and young adults. The sample comprised 30.5 white males, 32.4 white females, 17.5 black males and 19.6% black females.

2.2. General examination Identical protocols were used in all surveys [13]. Blood was drawn between 08:00 and 09:00 from participants who were instructed to fast for 12 h before the screening. Height was measured to within 0.1 cm, and weight to within 0.1 kg. As a measure of obesity, the body mass index (BMI =weight in kilograms divided by height in meters squared) was calculated. In 5–17 year olds, assessment of pubertal development was based on female breast or male genitalia appearance, determined by a physician, according to the ratings of Tanner [14]. Children were divided into five stages, ranging from prepubescence (stage 1) to complete development (stage 5). Systolic blood pressure was recorded at the first Korotkoff phase and diastolic at fourth and fifth phases. The fourth phase was used in children, the fifth phase in adults. The blood pressure levels reported were the mean of six replicate readings taken by two randomly assigned nurses. Parental history of cardiovascular disease was obtained through a questionnaire administered before the date of examination. For school aged children, the child’s parent or guardian completed the questionnaire. Young adults filled out the questionnaire by themselves. The respondents were asked to record whether the identified subject’s blood relative, mother and/or father, had ever had a heart attack, high blood pressure or stroke. No attempt was made to validate the parental history information. Further, information on age of onset of parental disease was not obtained. Since the study population is 5 – 30 years of age with 72% below the age of 25 years the majority of the reported parental history of disease is likely to be early onset in nature.

2.3. Laboratory analyses Serum total bilirubin levels were determined by the diazo method as part of the clinical chemistry measurements with a multichannel Olympus Au-5000 Analyzer (Olympus, Lake Success, NY).

Serum total cholesterol and triglycerides were measured by use of chemical procedures [15] on a Technicon AutoAnalyzer II (Technicon Instrument, Tarrytown, NY) until 1986. Since then these variables were determined by enzymatic procedures [16,17] on an Abbott VP instrument (Abbott Laboratories, North Chicago, IL). Both chemical and enzymatic procedures met the performance requirements of the Lipid Standardization Program of the Centers for Disease Control and Prevention, Atlanta, GA. Serum lipoprotein cholesterols were measured by a combination of heparin–calcium precipitation and agar–agarose gel electrophoresis procedures [18]. Plasma glucose levels were measured by an enzymatic method using the Beckman Instant Glucose Analyzer (Beckman Instruments, Palo Alto, CA). A commercial radioimmunoassay kit was used for measuring plasma immunoreactive insulin levels (Phadebas Pharmacia, Piscataway, NY). To assess the reproducibility of the entire process from blood sampling to data processing a second blood sample (blind duplicate) was collected on each screening day of a survey from an approximate 10% random subsample of participants. The intraclass correlation coefficients (a measure of reproducibility) between the blind duplicate values ranged from 0.93 to 0.99 for total cholesterol, 0.97 to 0.99 for triglycerides, 0.85 to 0.98 for VLDL cholesterol, 0.95 to 0.98 for LDL cholesterol, 0.92 to 0.95 for HDL cholesterol, 0.91 to 0.98 for glucose, 0.86 to 0.95 for insulin and 0.89 to 0.96 for bilirubin.

2.4. Statistical analyses Since the distribution of serum bilirubin was skewed, the values were transformed logarithmically for all tests of statistical significance. Log transformation was used as it provided the best transformation that provided a normal distribution. Since age-related changes were not uniform, subjects were grouped into specific age groups to reflect developmental periods of preadolescence, adolescence, and transition into adulthood. Race and sex differences in serum bilirubin levels were examined by an analysis of variance model which included race and sex main effects and a race-by-sex interaction. Significant predictors of serum bilirubin levels were explored using a stepwise regression procedure, with the significance levels for entry and staying specified at 0.50 and 0.05, respectively. The independent variables included age, race (0=white, 1= black), sex (0= male, 1=female), BMI, and, for ages 11–30 years, smoking (0= no, 1=yes) and alcohol use (0= no, 1=yes). Spearman partial correlation coefficients were used to examine the association of serum bilirubin levels with cardiovascular risk factor variables, partialing out the effects of age, BMI, smoking and alcohol intake. Mean

M. Madha6an et al. / Atherosclerosis 131 (1997) 107–113

109

Table 1 Serum bilirubin levels by age, race and sex Age group (years)

5– 10 11 – 17 18 – 24 25 – 30

White males

0.469 0.21 (n= 458) 0.59 9 0.29 (n=347) 0.759 0.46 (n= 235) 0.749 0.34 (n= 227)

Black males

0.439 0.22 (n =279) 0.5590.33 (n= 234) 0.6690.36 (n=122) 0.669 0.38 (n=92)

White females

0.48 90.24 (n =432) 0.53 9 0.28 (n = 353) 0.50 9 0.25 (n =297) 0.53 90.28 (n =266)

Black females

0.45 9 0.32 (n =276) 0.48 90.34 (n =225) 0.45 9 0.21 (n = 181) 0.47 90.22 (132)

P Race

Sex

0.0001

0.79

0.0001

0.0001

0.0001

0.0001

0.0001

0.0001

Results in mg/dl (mean 9S.D.).

levels of serum bilirubin were compared between offspring with and without a parental history of cardiovascular disease by an analysis of covariance, adjusting for appropriate covariates listed above. All analyses were performed using the Statistical Analysis System (SAS) software [19].

3. Results

3.1. Le6els by race, sex, and age Race- and sex-specific mean levels and selected percentiles of serum bilirubin are given in Tables 1 and 2, respectively, in the age range of 5 – 30 years as four age groups. The mean and percentile values showed a skewed distribution towards higher values in all race, sex, and age groups. Males and females alike showed a significant race difference (whites \blacks) in bilirubin levels within each age group. There was a significant sex difference, except in the age group of 5 – 10 years, with males of both races showing higher values than their female counterparts from the same age group. In males of both races the levels increased between the age groups 5–10 years and 18 – 24 years, and thereafter there was no change. In general, the age-related changes were less conspicuous in females although there was a marginal increase with age in white females. Pubertal development, as measured by Tanner stage in 5–17 year olds, showed no significant change in bilirubin levels among females of both races (data not shown). On the other hand, white males with Tanner stages 1–3 had lower values (P B0.05) than those with Tanner stages 4– 5. Black males with Tanner stages 1–4 had lower values (P B 0.05) than those with Tanner stage 5.

3.2. Predicltor 6ariables The independent relation of age, race, sex, adiposity (BMI), cigarette smoking, and alcohol use to serum bilirubin in different age groups was examined by a stepwise regression procedure (Table 3). All variables retained in the model were significant at PB0.05. Race was retained as a predictor variable in all age groups; sex and BMI in all age groups except the 5–10 years group; age in 5–10 years and 11–17 years group; and cigarette smoking in 18–24 years and 25–30 years groups. These variables accounted for 3–5% and 14– 17% of the variance in serum bilirubin in children and adults, respectively. It should be noted that both BMI and cigarette smoking associated inversely with bilirubin.

3.3. Relation to selected cardio6ascular risk factor 6ariables Partial Spearman correlations, controlled for age, BMI, cigarette smoking and alcohol use, between serum bilirubin and selected traditional physiologic cardiovascular risk factor variables are given in Table 4 for children and in Table 5 for adults by race and sex. In general, inverse correlation of varying statistical significance was noted for total cholesterol, VLDL cholesterol, triglycerides, LDL cholesterol, insulin, glucose, and systolic blood pressure in certain age-race-sex groups; in contrast, HDL cholesterol showed a positive association of varying significance. Within the age group no association was noted between bilirubin and diastolic blood pressure, whether recorded at the fourth or the fifth Korotkoff phase. Among the age-race-sex groups black males aged 18–30 years showed no significant correlation for any of these variables.

110

M. Madha6an et al. / Atherosclerosis 131 (1997) 107–113

Table 2 Selected percentiles of serum bilirubin by age, race and sex Age group (years)/percetiles

White males

Black males

White females

Black females

5 – 10 (n) 5th 50th 95th

458 0.2 0.4 0.9

279 0.2 0.4 0.8

432 0.2 0.4 0.9

276 0.2 0.4 0.9

11 – 17 (n) 5th 50th 95th

347 0.3 0.5 1.1

234 0.3 0.5 1.0

353 0.2 0.5 1.0

225 0.2 0.4 1.0

18 – 26 (n) 5th 50th 95th

235 0.3 0.7 1.4

122 0.3 0.6 1.4

297 0.2 0.5 1.0

181 0.2 0.4 0.8

25 – 30 (n) 5th 50th 95th

227 0.3 0.7 1.5

92 0.2 0.6 1.3

266 0.2 0.5 1.0

132 0.2 0.4 0.9

3.4. Relation to parental history of cardio6ascular disease The relation of serum bilirubin levels to cardiovascular risk was further examined in terms of parental history of cardiovascular disease, a surrogate measure of the offspring’s risk. Mean levels of serum bilirubin (adjusted for age, race, sex, BMI, cigarette smoking, and alcohol use) are shown in Table 6, according to parental history of cardiovascular disease. Individuals with parental history of heart attack had decreased levels of bilirubin compared with those without such parental history (P = 0.06). Positive parental history of hypertension was also associated with decreased levels of serum bilirubin in the offspring (P =0.01). Although the mean bilirubin level tended to be lower in individuals with parental history of stroke than those without such parental history, the difference was not statistically significant.

4. Discussion Bilirubin has long been considered as a waste product of heme metabolism with no apparent pathophysiologic function except that high concentrations in serum serves as a marker for certain clinical conditions especially hepatobiliary disorders. However, recent studies indicate that bilirubin serves as an endogenous tissue protector by attenuating radical-mediated damage to both lipids and proteins [20]. In fact, under low oxygen concentrations the antioxidant activity of bilirubin surpasses that of a-tocopherol [2]. In addition to being a potent antioxidant, bilirubin is reported to play a role in tissue protection against inflammatory damage

by its anti-complement action [8]. It has been reported that albumin, which appears in inflammatory exudate, carries bilirubin across the vascular wall into the sites of potential oxyradical damage by phagocytic cells [21,22]. Low-density lipoprotein lipid oxidation-mediated inflammatory response in the artery is considered crucial to atherosclerotic lesion development [1,23,24]. Further, there is evidence that hypertension may also exert oxidative stress on the arterial wall to activate an inflammatory response [25]. Thus, bilirubin may have a beneficial role in atherogenesis. Viewed in this context it is of interest that our results show that the levels of serum bilirubin in the offspring, which are race-, sex- and age-specific, relate inversely to parental history of heart attack and hypertension. It should be noted that these observations are derived from free-living children and young adults from a total biracial (black–white) community study. To our knowledge this is the first report showing black–white differences in bilirubin levels, and linking bilirubin levels to the familial association of cardiovascular risk. Before discussing our results it should be noted that since the present study was not designed to address the issue of underlying mechanisms, the explanations provided are tentative. The current finding that blacks of both sexes showed lower levels of bilirubin than their white counterparts may be related to black–white difference in hemoglobin levels, since bilirubin is a catabolic product of heme. Earlier studies including our own have shown consistently higher hemoglobin in whites versus blacks, independent of socioeconomic status and dietary iron intake [26–28]. Further, the sex- and age-related differences in bilirubin levels also parallel the corresponding differences in hemoglobin reported previously [26–28].

M. Madha6an et al. / Atherosclerosis 131 (1997) 107–113

111

Table 3 Predictor variables of serum bilirubin in children and young adults Age group (years)

n

Predictor variable

Regression coefficient (S.E.)

5– 10

1445

11 – 17

1159

18 – 24

835

25 – 30

717

Race Age Age Sex Race BMI Sex Smoking Race BMI Sex Race Smoking BMI

−0.098 0.035 0.035 −0.119 −0.110 −0.006 −0.371 −0.162 −0.139 −0.011 −0.345 −0.112 −0.120 −0.006

(0.024) (0.007) (0.007) (0.029) (0.030) (0.003) (0.035) (0.037) (0.036) (0.003) (0.035) (0.038) (0.036) (0.003)

Partial R 2 0.011 0.015 0.020 0.015 0.012 0.003 0.121 0.014 0.018 0.013 0.113 0.014 0.013 0.004

Independent variables: age; race (0 = white, 1 =black); sex (0 =male, 1 =female); BMI; smoking (0 =No, 1 =Yes, 11 – 30 years only) and alcohol intake (0= No, 1 = Yes, 11–30 years only).

Bilirubin levels were higher in males of both races than their female counterparts, except in the preadolescent age group of 5 – 10 years. However, males in general have higher risk for cardiovascular disease than females. This apparent paradox may be due to the multifactorial nature of cardiovascular disease. For example, while increased levels of bilirubin may be protective in males, increased levels of antiatherogenic HDL cholesterol and estrogen may counteract the effect of decreased levels of bilirubin and confer even better protection in premenopausal females. Our results show that in addition to race, sex, and age, adiposity and cigarette smoking also contributed to the variability of bilirubin in serum. The inverse association of body fatness with serum bilirubin has not been reported previously. With respect to cigarette smoking, which is known to generate oxygen and carbon-centered radicals [29], an earlier study also has shown that smoking significantly lowers serum bilirubin

levels [30]. Although alcohol intake was not associated with bilirubin in the present study, earlier studies have shown either a positive association [30] or no association [31]. In the present study bilirubin levels tended to be low in offspring of parents who reported having had a heart attack (P= 0.06) or hypertension (P = 0.01) compared with those without a positive parental history, independent of age, race, sex, cigarette smoking, and alcohol use. Although the relationship of bilirubin to parental history of heart attack is of borderline significance, this association is in the same direction noted for parental history of hypertension or stroke. Therefore, it is reasonable to exclude chance as a reason for the observed relationship. Self-reported parental histories were not verified in the present study. Others have found 73– 81% concordance in the case of self-reported myocardial infarction [32,33]. The present finding is noteworthy, since nonsystematic misclassification by

Table 4 Correlation of serum bilirubin to selected cardiovascular risk factor variables in 5 to 17-year-olds by race and sex Risk factor variable

Cholesterol Total LDL VLDL HDL Triglycerides Insulin Glucose Systolic blood pressure Diastolic blood pressure (4th phase)

Males

Females

White (n =782)

Black (n =501)

White (n = 761)

Black (n = 486)

0.03† 0.02 −0.12*** 0.15**** −0.18**** −0.21**** 0.02 0.00 0.00

−0.10* −0.11* −0.08 0.04 −0.16*** −0.17**** −0.05 −0.02 −0.02

0.00 −0.02 −0.10** 0.12*** −0.15**** −0.08* 0.05 −0.01 0.04

−0.08 −0.03 −0.04 0.10* −0.08 −0.03 −0.10* −0.01 −0.08

†Partial Spearman correlation coefficients, adjusted for the effects of age and BMI. * PB0.05; ** PB0.01; *** PB0.001; **** PB0.0001.

M. Madha6an et al. / Atherosclerosis 131 (1997) 107–113

112

Table 5 Correlation of serum bilirubin to selected cardiovascular risk factor variables in 18 to 30-year-olds by race and sex Variable

Males

Cholesterol Total LDL VLDL HDL Triglycerides Insulin Glucose Systolic blood pressure Diastolic blood pressure (5th phase)

Females

White (n =421)

Black (n =176)

White (n = 539)

Black (n =287)

−0.07† −0.07 −0.12* 0.15** −0.20**** −0.15** −0.20**** −0.03 0.02

0.12 0.12 −0.03 0.04 −0.04 0.01 −0.13 −0.09 −0.03

−0.14*** −0.15*** −0.02 0.02 −0.09* −0.10a 0.01 −0.10* 0.04

−0.02 −0.10 −0.09 0.12* −0.11 −0.11 0.01 0.03 0.11

†Partial Spearman correlation coefficients, adjusted for age, BMI, smoking and alcohol intake. * PB0.05; ** PB0.01; *** PB0.001; **** PB0.0001.

self-reported histories generally underestimate the risk potential. Within the normal range of the serum bilirubin low levels may be a marker for oxidative stress and future cardiovascular disease. However, whether bilirubin level per se is familial in nature or bilirubin is a marker for some heritable or metabolic trait(s) related causally to cardiovascular disease is not clear. Recent findings [34] that higher serum bilirubin is associated with decreased risk for early familial CAD seem to support the familial concept. Further, evidence for a major gene elevating serum bilirubin concentrations, which may be protective against CAD, has been presented [35]. The relationship of serum bilirubin to traditional physiologic cardiovascular risk factor variables in this study population is of interest. For example, bilirubin showed positive association with HDL cholesterol and an inverse association with adiposity, LDL cholesterol, systolic blood pressure, insulin and glucose. Current findings give credence to the novel finding by Schwertner et al. [9] who found that serum bilirubin is an independent inverse risk factor for angiographically Table 6 Serum bilirubin levels of offspring with a parental history of cardiovascular disease Parental disease Heart attack No (n = 2149) Yes (n =363) Hypertension No (n = 1367) Yes (n =1178) Stroke No (n =2323) Yes (n =148) a

Bilirubin (mg/dl)

No vs. Yes (P)

0.559 0.01a 0.5290.02

0.06

0.579 0.01 0.5390.01

0.01

0.55 90.01 0.53 90.02

0.19

Adjusted for age, race, sex, BMI, cigarette smoking and alcohol intake (mean9 S.E.).

documented coronary artery disease, with an association equivalent in degree to that of systolic blood pressure. Further studies are needed to establish the relationship between bilirubin and cardiovascular disease in other populations and to evaluate the underlying mechanisms involved in this relationship.

Acknowledgements The Bogalusa Heart Study represents the collaborative efforts of many people whose cooperation is gratefully acknowledged. We especially thank Bettye Seal for her work as Community Coordinator. We also thank the children and young adults of Bogalusa without whom this study would not be possible. We express appreciation to Dr H.A. Schwertner for suggesting these analyses. This study was supported by grants HL38844 from the National Heart, Lung, and Blood Institute and HD32194 from the National Institute of Child Health and Human Development of the U.S. Public Health Service. Dr Madhavan is a recipient of an International Atherosclerosis Society Visiting Fellowship.

References [1] Steinberg D, Parthasarathy S, Carew TE, Khoo JC, Witzum JL. Beyond cholesterol. Modification of low-density lipoprotein that increases its atherogenicity. N Engl J Med 1989;320:915–925. [2] Stocker R, Yamamoto Y, McDonagh AF, Glazer AN, Ames BN. Bilirubin is an antioxidant of possible physiologic importance. Science 1987;235:1043 – 1046. [3] Wu TW, Carey D, WU J. The cytoprotective effects of bilirubin and biliverdin on rat hepatocytes and human erythrocytes and the impact of albumin. Biochem Cell Biol 1991;69:828–834. [4] Wu TW, Wu J, Li RK, Mickle D, Carey D. Albumin-bound bilirubins protect human ventricular myocytes against oxyradical damage. Biochem Cell Biol 1991;69:683 – 688.

M. Madha6an et al. / Atherosclerosis 131 (1997) 107–113 [5] Wu TW, Fung KP, Yang CC, Unconjugated bilirubin inhibits the oxidation of human low density lipoprotein better than Trolox. Life Sci 1994;54:477–481. [6] Llesuy SF, Tomaro ML. Heme oxygenase and oxidative stress. Evidence of involvement of bilirubin as physiological protector against oxidative damage. Biochim Biophs Acta 1994;1223:9 – 14. [7] Hegyi T, Goldie E, Hiatt M. The protective role of bilirubin in oxygen-radical diseases of the preterm infant. J Perinatol 1994;14:296 – 300. [8] Nakagami T, Toyomura K, Kinoshita T, Morisawa S. A beneficial role of bile pigments as an endogenous tissue protector: anticomplement effects of biliverdin and conjugated bilirubin. Biochim Biophys Acta 1993;1158:189–193. [9] Schwertner HA, Jackson WG, Tolan G. Association of low serum concentration of bilirubin with increased risk of coronary artery disease. Clin Chem 1994;40:18–23. [10] ten Kate LP, Bowman H, Dalger SP, Motulsky AG. Familial aggregation of heart disease and its relation to known risk factors. Am J Cardiol 1982;50:945–953. [11] Moll PP, Sing CF, Weidman WH et al. Total cholesterol in school children: predictions of coronary heart disease in relatives. Circulation 1983;67:127–134. [12] Freedman DS, Srinivasan SR, Shear CL, Franklin FA, Webber LS, Berenson GS. The relation of apolipoproteins A-I and B in children to parental myocardial infarction: The Bogalusa Heart Study. N Engl J Med 1986;315:721–726. [13] Berenson GS, McMahan CA, Voors AW et al. Cardiovascular Risk Factors in Children—The Early Natural History of Atherosclerosis and Essential Hypertension. New York: Oxford University Press, 1980;450. [14] Tanner JM. Growth at Adolescence, 2nd ed. Oxford: Blackwell Scientific Publications, 1962;326. [15] Lipid Research Clinics Program Manual of Laboratory Operations, Vol 1: Lipid and Lipoprotein Analysis. U.S. Department of Health, Education, and Welfare. publication no. (NIH) 75628, National Institutes of Health, 1974. [16] Allain CC, Poon LS, Chan CSG, Richmond W, Fu PC. Enzymatic determination of total serum cholesterol. Clin Chem 1974;20:470 – 475. [17] Bucolo G, David H. Quantitative determination of serum triglycerides by the use of enzymes. Clin Chem 1973;19:476 – 482. [18] Srinivasan SR, Berenson GS. Serum lipoproteins in children and methods for study. In: Lewis LA (ed). CRC Handbook of Electrophoresis, Vol 3. Lipoprotein Methodology and Human Studies. Boca Raton, FL: CRC Press, 1983;184–204. [19] SAS User’s Guide: Statistics, Version 6. Cary, NC: SAS Institute, 1989. [20] Neuzil J, Stocker R. Bilirubin attenuates radial-mediated damage to serum albumin. FEBS Let 1993;331:281–284.

.

113

[21] To¨gl-Leimuller A, Egger G, Porta S. Albumin as one-way transport vehicle into sites of inflammation. Exp Pathol 1986;30:91– 96. [22] Fantone JC, Ward PA. Role of oxygen-derived free radicals and metabolites in leukocyte-dependent inflammatory reactions. Am J Pathol 1982;107:395 – 418. [23] Berliner JA, Territo MC, Sevanian A et al. Minimally modified low density lipoprotein stimulates monocyte endothelial interactions. J Clin Invest 1990;85:1260 – 1266. [24] Rajavashisth TB, Andalibi A, Territo MC et al. Induction of endothelial cell expression of granulocyte and macrophage colony-stimulating factors by modified low-density lipoproteins. Nature 1990;344:254 – 257. [25] Alexander RW. Hypertension and the pathogenesis of atherosclerosis. Oxidative stress and the mediation of arterial inflammatory response: a new perspective. Hypertension 1995;25:155 – 161. [26] Frerichs RR, Webber LS, Srinivasan SR, Berenson GS. Hemoglobin levels in children from a biracial southern community. Am J Public Health 1977;67:841 – 845. [27] Voors AW, Frank GC, Srinivasan SR, Webber LS, Berenson GS. Hemoglobin levels and dietary iron in pubescent children in a biracial community. Public Health Rep 1981;96:45–49. [28] Garn SM, Smith NJ, Clark DC. Lifelong differences in hemoglobin levels between blacks and whites. JAMA 1975;67:91 – 96. [29] Church DF, Pryor WA. Free-radical chemistry of cigarette smoke and its toxicological implications. Environ Health Perspect 1985;64:111 – 126. [30] Chan-Yeung M, Ferreira P, Frohlich J, Schulzev M, Tan F. The effects of age, smoking, and alcohol in routine laboratory tests. Am J Clin Pathol 1981;75:320 – 326. [31] Shaper AG, Pocock SJ, Ashby D, Walker M, Whitehead TP. Biochemical and haematological response to alcohol intake. Ann Clin Biochem 1985;22:50 – 61. [32] Napier JA, Metzner H, Johnson BC. Limitations of morbidity and mortality data obtained from family histories: a report from the Tecumseh Community Health Study. Am J Public Health 1972;62:30 – 35. [33] Tretli S, Lund-Larsen PG, Foss OP. Reliability of questionnaire information of cardiovascular disease and diabetes: Cardiovascular disease study in Finnmarle County. J Epidemiol Community Health 1982;36:269 – 273. [34] 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. [35] 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.