a University of Leuven, Department of Public Health, Division of Nutritional Epidemiology, Kapucijnenvoer 3335, B3000 Leuven, Belgium.
b Wageningen University, Human Nutrition and Epidemiology, PO Box 8129, 6700 EV Wageningen, The Netherlands.
Pascale GA Van Hoydonck, Department of Public Health, Division of Nutritional Epidemiology, Kapucijnenvoer 35, B-3000 Leuven, Belgium. E-mail: Pascale.Vanhoydonck{at}med.kuleuven.ac.be
Abstract
Background The endogenous antioxidant serum bilirubin may scavenge free radical species from cigarette smoke. Smokers are expected to have lower serum bilirubin levels than never smokers, but this has never been evaluated in a random population sample of women and men. In addition, in vitro studies indicate that filter cigarette smoke might enhance bilirubin oxidation more than non-filter cigarette smoke.
Methods In this paper, data were analysed from the Belgian Interuniversity Research on Nutrition and Health (BIRNH) study. We investigated the association of smoking and type of cigarettes with serum bilirubin concentrations in men and women using multiple regression models taking into account differences in baseline and smoking characteristics.
Results In men, current smokers had lower crude and adjusted serum bilirubin concentrations compared with never smokers (P = 0.0001). For women, the association between smoking and serum bilirubin concentrations adjusted for age was in the same direction as for men, but did not reach statistical significance. Male former smokers had serum bilirubin concentrations in between those of current smokers and never smokers, while female former smokers had even higher (+ 0.50 µmol/l) serum bilirubin concentrations than never smokers. Male subjects smoking filter cigarettes showed a higher serum bilirubin concentration than those smoking cigarettes without a filter (adjusted difference: 0.49 µmol/l; P = 0.03). The duration of smoking was inversely and significantly related to serum bilirubin concentrations (P = 0.0003).
Conclusion Smoking may lower serum bilirubin concentrations in men, especially in those smoking non-filter cigarettes. The duration of smoking is a more important determinant for serum bilirubin concentrations than the number of cigarettes smoked per day, which may indicate the cumulative negative effects of smoking on the endogenous antioxidant system. Results for females pointed in the same direction, but were less clear, possibly due to low numbers of smoking women, who were largely young and smoked mainly filter cigarettes for a shorter period of time.
Keywords Bilirubin, smoking status, type of cigarettes
Accepted 4 January 2001
Elevated bilirubin concentration is regarded as a useful marker of liver and blood disorders, and of metabolic alterations. Bilirubin has also been reported to have antioxidant activity.17 Recent unpublished results from our group indicated an inverse association of serum bilirubin concentration with cancer mortality.
The frequently quoted reference interval of total serum bilirubin concentrations in healthy adults is 3.4217.10 µmol/l (0.21.0 mg/dl). In general, 17.10 µmol/l (1 mg/dl) is accepted as the upper limit of normal. The concentration of serum bilirubin is higher in men than in women and decreases with age.810 Other determinants of serum bilirubin concentrations include smoking status,8,1114 fasting status,9 and current oral contraceptive use in women.8 There is a considerable within-subject variation and a variation coefficient of 25% was reported for repeated serum bilirubin concentration measurement covering a period of 2 weeks at intervals of 23 days between blood sampling.15
Cigarette smoke contains a large variety of compounds, including many oxidants and free radicals that are capable of causing a prooxidant/antioxidant imbalance in the blood and tissues of smokers.1618 Pryor et al.19 have identified two different populations of free radicals, one in the tar and one in the gas phase, in cigarette smoke.
The association of tobacco smoking and serum bilirubin concentrations has been described in some studies1113 which have shown lower serum bilirubin concentrations in male smokers compared to male non-smokers. However, no results have been reported on female smokers. In vitro studies showed that exposure to whole ( non-filter) cigarette smoke, in contrast to filtered cigarette smoke, did not induce detectable lipid peroxidation or bilirubin oxidation because of the antioxidant properties of the tar compound (melanin-like phenolic pigments).20,21 Up till now variations in serum bilirubin concentrations of subjects smoking cigarettes with or without filters have never been evaluated in vivo.
In this paper we examine the association between tobacco smoking and serum bilirubin concentrations in men as well as in women. We also study serum bilirubin concentrations in smokers smoking different types of cigarettes.
Methods
Study population
The Belgian Interuniversity Research on Nutrition and Health (BIRNH study) was carried out from 1980 to 1984. In each of the 42 counties of Belgium, three to five municipalities were randomly chosen and the most populated was always included. In the selected municipalities a random sample of the population was drawn from the voting list. Of the Belgians invited to participate, 38.6% of the men (n = 5949) and 34.4% of the women (n = 5353) aged 2574 years volunteered for the study.
Measurements
All subjects filled out a standardized questionnaire which included questions about age, educational level, smoking, menopausal status, use of oral contraceptives and a pre-structured 24-hour dietary record (which was fully reviewed by a trained dietician). The smoking questionnaire contained detailed information on history of tobacco smoking (average number of cigarettes per day, number of years of smoking, age at which they began and/or stopped smoking, use of cigarettes with or without filter or handrolled cigarettes). Blood and urine were collected and analysed for a number of biochemical parameters. The aims, design and methodology of the BIRNH study have been described previously.22
Serum bilirubin concentration was measured on a Technicon Autoanalyzer II continuous flow analytical instrument that utilizes the automated method of Gambino and Freda,23 based on the method of Jendrassik and Grof. This method uses a blank channel, which runs together with the sample channel to correct for endogenous factors in the sample interfering with total bilirubin concentrations. Serum bilirubin concentrations are given in µmol/l (mg/dl * 17.10).
Statistical analysis
Statistical analysis of the data was carried out using the statistical software programme SAS® version 6.12 (SAS Institute, Inc., Cary, North Carolina).
Data from 892 people were excluded from the statistical analysis due to missing information leaving 10 400 subjects (5524 men and 4876 women).
Baseline characteristics of the participants were classified according to smoking status (never, current, former) and for male current smokers according to smoking type (non-filter, filter, handrolled and combinations of cigarettes). Data from female smokers were not analysed according to smoking type because of low numbers in three of the four groups. Smoking any type of tobacco was defined as current smokers. Educational level was classified into three groups: low (up to completed primary school), intermediate (high school) and high (professional higher education or university). Alcohol consumption (daily energy per cent [En%]) was classified into three groups: 0, 4 and >4.
The number of handrolled cigarettes smoked a day was calculated as g tobacco/day * 0.8. It was assumed that one package contains 50 g and 40 cigarettes can be obtained per package. The number of pack-years was calculated from the number of cigarettes per day divided by 25 (number of cigarettes per package) and multiplied by the years of smoking.
Differences in baseline characteristics and serum bilirubin concentrations between the categories of smoking status and smoking type were tested with analysis of variance (ANOVA), using Scheffé test for multiple comparisons. Differences in categorical variables (educational level, alcohol intake) were tested with 2 tests. In multiple regression models, differences in serum bilirubin concentrations according to smoking status and type were estimated adjusting for possible confounders. Data on smoking status, smoking type, educational level and alcohol intake were entered into the models as dummy variables. The distribution of serum bilirubin concentrations in men as well as in women was skewed to the right. Therefore, all analyses were repeated with log-transformed serum bilirubin data. A P-value of <0.05 was regarded as statistically significant and all P-values were based on two-sided tests.
Results
Smoking status
The baseline characteristics according to categories of smoking status in men (n = 5459) and women (n = 4876) are shown in Tables 1 and 2 respectively. The prevalence of current smoking in males/females was: 48%/16%; former smoking: 29%/9%; non-smoking: 22%/73%.
|
|
Figure 1 shows the age-adjusted means of serum bilirubin concentrations in men and women. Male current smokers had significantly lower serum bilirubin concentrations (7.04 µmol/l) than never (8.32 µmol/l) (P < 0.0001) and former smokers (7.99 µmol/l) (P < 0.05). The results did not change for male current smokers versus never smokers and were slightly less pronounced for smokers versus former smokers (Table 4a
) when adjustments were made for other differences in baseline variables (age, educational level and iron intake). Parameter estimates ± standard error in men for the covariables adjusted for in the full model were 0.025 ± 0.004 (P = 0.0001) for age (years), 0.212 ± 0.126 (P = 0.0912) for intermediate versus low educational level, 0.599 ± 0.151 (P = 0.0001) for high versus low educational level and 0.040 ± 0.010 (P = 0.0001) for iron intake (mg/d).
|
|
Parameter estimates ± standard error in women for the covariables adjusted for in the full model were 0.008 ± 0.004 (P = 0.060) for age (years), 0.250 ± 0.107 (P = 0.0188) for intermediate versus low educational level, 0.251 ± 0.153 (P = 0.1009) for high versus low educational level, 0.048 ± 0.011 (P = 0.0001) for iron intake (mg/d) and 0.015 ± 0.010 (P = 0.1314) for PUFA intake (En%).
Adding actual use of contraceptives in the model the results remained essentially unchanged.
Smoking type
Baseline characteristics according to categories of smoking type in men (n = 2233) are shown in Table 3. The filter cigarette smokers constituted the largest group (39%) of male current smokers, followed by handrolled cigarette smokers (32%), combination of different types of cigarette smokers (14%) and finally the non-filter cigarette smokers group (14%). Most female current smokers smoked filter cigarettes (89%).
|
Male subjects smoking filter cigarettes had a higher (+ 0.73 µmol/l) serum bilirubin concentration than those smoking cigarettes without a filter (P = 0.0009) (Figure 2 and Table 4b
). Smokers of handrolled cigarettes and of combinations of different types of cigarettes presented a higher, but not significant, serum bilirubin concentration compared to the reference group, non-filter cigarette smokers. The results did not change when adjustments were made for age, duration of smoking, BMI, educational level and iron intake in male cigarette smokers. The duration of smoking was inversely and significantly related to serum bilirubin concentrations independently of age (Figure 3
) and was the only smoking variable determining the variations in serum bilirubin. The number of cigarettes smoked per day was not associated with serum bilirubin concentration.
|
|
|
Discussion
In this study serum bilirubin concentrations were significantly lower in male smokers compared with never smokers, while intermediate concentrations were observed for former smokers. Lower serum bilirubin was also found in female smokers compared with never smokers only after adjustment for age, however it was not significant. Serum bilirubin concentrations of female former smokers were significantly higher than those of never smokers. Among the male smokers, those smoking filter cigarettes had the highest serum bilirubin concentrations followed by smokers of handrolled and of non-filter cigarettes.
The low serum bilirubin concentrations of male Belgian smokers confirmed previous observations,1114 reported new results for women and adjusted for possible confounders. Bilirubin, which is a bile pigment, is generally regarded as waste product of haem catabolism. However, it has also been suggested that bilirubin may have an important role as an endogenous antioxidant and is destroyed by reactive oxygen species.17 The observed low bilirubin concentrations in males and, when adjusted for age, in female smokers might be the result of an overconsumption of bilirubin by free radical species related to cigarette smoke.21 Another possible mechanism which can determine serum bilirubin concentration is hepatic bilirubin UDP-glucuronyltransferase (UGT) activity.24 Cigarette smoke is known to induce UGT-activity25 by which hepatic bilirubin UGT is inversely correlated with serum bilirubin concentration.
The relation between serum bilirubin and smoking status was controlled for differences in age, educational level and dietary intake of prooxidants. Higher serum bilirubin concentrations were observed in younger people (this study and ref. 10) and with lower intakes of prooxidants. In addition, educational level was different among smoking classes and associated with serum bilirubin. After controlling for age in females, the association between bilirubin and smoking status was in the same direction as for men.
The higher serum bilirubin concentration in male former smokers compared to current smokers, as others have also observed,10,11 may indicate that bilirubin concentrations are restored after a period of high free radical exposure due to tobacco smoke. The higher serum bilirubin concentrations in female former smokers compared with never smokers may be in part due to changes in dietary habits. Diets of female former smokers contained higher amounts of some antioxidants compared with female never smokers, while the dietary habits of male former smokers resembled more those of never smokers rather than current smokers (this study and refs 26,27). In vitro studies have indicated that serum vitamin C4 lowers bilirubin oxidation, and lipid peroxides from PUFA2 or iron28 enhance bilirubin oxidation. The association between these nutrient intakes and bilirubin concentrations, however, has never been evaluated in human experimental studies. In the BIRNH study dataset, we have observed differences in intakes of prooxidants rather than antioxidants with varying serum bilirubin concentrations. Diets of men and women with the highest serum bilirubin concentrations contained less iron and less PUFA than the diets of those with the lowest serum bilirubin concentrations. In the present study, adjustments for differences in dietary intake of iron and PUFA (women only) did not change the association between bilirubin levels and smoking status in men and women.
The second aim of this study was to investigate the type of cigarette smoked in relation to serum bilirubin concentrations. This was not possible in female smokers because of the low numbers of women smoking non-filter cigarettes. Higher serum bilirubin concentrations were observed in male subjects smoking filter cigarettes compared to smokers of cigarettes without a filter. Previous in vitro studies20,21,29 showed that exposure to whole (non-filter) cigarette smoke, in contrast to filter cigarette smoke, did not induce detectable lipid peroxidation or cause bilirubin oxidation once endogenous ascorbic acid has been oxidized completely. This may be due to the antioxidant properties of the cigarette smoke tar compound (melanin-like phenolic pigments) which counteracts the prooxidant capacity of aqueous tar extracts. Furthermore, the gas phase of cigarette smoke contains small oxygen and carbon-centred radicals that are much more reactive than are the tar phase radicals.19,30 Serum bilirubin concentrations in smokers of filter and non-filter cigarettes have never been evaluated in vivo. Obviously, when studying in vivo filter versus non-filter cigarette smokers in relation to serum bilirubin concentration many other important factors influence this association. Our data showed that filter cigarette smokers have other baseline characteristics compared to smokers of non-filter cigarettes which could bias the results. They were 5 years younger and had higher education than smokers of other types of cigarettes. Filter cigarette smokers, in addition, smoked for a shorter time and fewer cigarettes per day than non-filter cigarette smokers. Our data indicate that the duration of smoking was a more important determinant of serum bilirubin concentrations than the number of cigarettes smoked per day. This suggests cumulative effects of smoking on the endogenous antioxidant system. It may well be that oxidants released as a result of chronic airway inflammation and oxidants in the smoke itself act synergistically. The serum bilirubin concentrations of the filter cigarette smokers were still higher than those of non-filter cigarette smokers, when adjusted for differences in baseline and smoking characteristics. The cigarette filter may have protective effects, which may explain the differences in serum bilirubin concentration between men smoking filter versus non-filter cigarettes. Factory-made cigarette filters, in addition, have different characteristics compared with Cambridge filters used in in vitro experiments on oxidants in cigarette smoke.1921,29,30
In this study, regression dilution bias can be expected because the blood sample for bilirubin measurement was taken only once. Winkel et al.15 evaluated the intra-individual variation of some serum constituents and observed an intra-subject coefficient of variation of 25% for bilirubin. Repeated measurements could, therefore, have improved measurement precision. For data on smoking characteristics we relied on self-reports of smoking, which is generally considered as a valid and accurate method.31,32 Differences in serum bilirubin concentration according to smoking status and type were estimated adjusting for age, duration of smoking, educational level and iron intake. It is unlikely that other determinants such as genetic factors, degree of light exposure and fasting status would have differed between smoking groups.
To summarize, smoking may lower serum bilirubin concentrations in men, especially in those smoking non-filter cigarettes. The duration of smoking is a more important determinant for serum bilirubin concentrations than the number of cigarettes smoked per day, which may indicate the cumulative negative effects of smoking on the endogenous antioxidant system. Results for females pointed in the same direction, but were less clear, possibly due to low numbers of smoking women, who were in majority young and smoked mainly filter cigarettes for a shorter period of time.
KEY MESSAGES
|
Acknowledgments
The BIRNH study was supported by the National Fund for Scientific Research grant no. 3.9002.79 and the Algemene Spaar-en Lijfrente kas (parastatal insurance company), Brussels, Belgium. The Unilever Chair in Nutritional Epidemiology, Catholic University of Leuven, Belgium, funded this research project. Participating universities and principal staff of the BIRNH study were: Gent State University: Dr Gaston Verdonk, Dr Karel Vuylsteek, Dr Guy De Backer, Greet Haelterman and Chris Seynaeve; Catholic University of Leuven: Dr Jozef V Joossens, Dr Hugo Kesteloot and Jef Gebroers; Free University of Brussels (ULB); Dr Marcel Graffar, Dr Marcel Kornitzer, Dr Claude Thilly, Werner Vanneste, Dr Michèle Dramaix, Francoise Kittel, Liliane Ravet, Anne van Hemeldock and Henri Darquennes; Free University of Brussels (VUB): Dr Anne-Marie Depoorter; Liège State University: Dr Gilberte Reginster-Haneuse; International Agency for Research on Cancer, Lyon, France: Dr Albert Tuyns.
References
1 Minetti M, Mallozzi C, Di Stasi AM, Pietraforte D. Bilirubin is an effective antioxidant of peroxynitrite-mediated protein oxidation in human blood plasma. Arch Biochem Biophys 1998;15:16574.
2 Stocker R, Glazer AN, Ames BN. Antioxidant activity of albumin-bound bilirubin. Proc Natl Acad Sci 1987;84:591822.[Abstract]
3 Stocker R, Yamamoto Y, McDonagh AF, Glazer AN, Ames BN. Bilirubin is an antioxidant of possible physiological importance. Science 1987;23:104346.
4 Yamaguchi T, Horio F, Hashizume T et al. Bilirubin is oxidized in rats treated with endototoxin and acts as a physiological antioxidant synergistically with ascorbic acid. Biochem Biophys Res Commun 1995; 214:1119.[ISI][Medline]
5 Yamaguchi T, Hashizume T, Tanaka M et al. Bilirubin oxidation provoked by endotoxin treatment is suppresses by feeding ascorbic acid in a rat mutant unable to synthesize ascorbic acid. Eur J Biochem 1997;15:23340.
6 Frei B, Stocker R, Ames BN. Antioxidant defenses and lipid peroxidation in human blood plasma. Med Sci 1988;85:974852.
7 Hidalgo FJ, Zamora R, Dillard CJ, Tappel AL. Can serum bilirubin be an index of in vivo oxidative stress? Med Hypothesis 1990;33:20711.[ISI][Medline]
8
Manolio TE, Burke GL, Savage PJ et al. Sex- and race-related differences in liver-associated serum chemistry tests in young adults in the CARDIA Study. Clin Chem 1992;38:185359.
9
White GL, Nelson JA, Pedersen DM, Ash KO. Fasting and gender (and altitude?) influence reference intervals for serum bilirubin in healthy adults. Clin Chem 1981;27:114042.
10 Hale WE, Stewart RB, Marks RG. Haematological and biochemical laboratory values in an ambulatory elderly population: an analysis of the effects of age, sex and drugs. Age Ageing 1983;12:27584.[Abstract]
11 Chan-Yeung M, Ferreira P, Frohlich J, Schulzer M, Tan F. The effects of age, smoking, and alcohol on routine laboratory tests. Am J Clin Pathol 1981;75:32026.[ISI][Medline]
12
Hopkins PN, Wu LL, Hunt SC. Higher serum bilirubin is associated with decreased risk for early familial coronary artery disease. Arterioscler Thromb Vasc Biol 1996;16:25055.
13 Schwertner HA. Association of smoking and low serum bilirubin antioxidant concentrations. Atherosclerosis 1998;136:38387.[ISI][Medline]
14
Schwertner HA, Jackson WG, Tolan G. Association of low serum concentration of bilirubin with increased risk of coronary artery disease. Clin Chem 1994;40:1823.
15
Winkel P, Statland BE, Bokelund H. Factors contributing to intra-individual variation of serum constituents: 5. Short-term day-to-day and within-hour variation of serum constituents in healthy subjects. Clin Chem 1974;20:152027.
16 Anderson R. Assessment of the roles of vitamin C, vitamin E, and beta-carotene in the modulation of oxidant stress mediated by cigarette smoke-activated phagocytes. Am J Clin Nutr 1991;53:358S61S.[Abstract]
17 Davies KJA. Oxidative stress: the paradox of aerobic life. Biochem Soc Symp 1995;61:131.[Medline]
18 Chow CK. Cigarette smoking and oxidative damage in the lung. Ann NY Acad Sci 1993;28:28998.
19 Pryor WA, Stone K. Oxidants in cigarette smoke: radicals, hydrogen peroxide, peroxynitrate, and peroxynitrite. Ann NY Acad Sci 1993;28:1228.
20 Lapenna D, De Gioia S, Mezzetti A et al. Cigarette smoke, ferritin, and lipid peroxidation. Am J Respir Crit Care Med 1995;151:43135.[Abstract]
21 Frei B, Forte TM, Ames BN, Cross CE. Gas phase oxidants of cigarette smoke induce lipid peroxidation and changes in lipoprotein properties in human blood plasma. Biochem J 1991;277:13338.[ISI][Medline]
22 De Backer G. Regional differences in dietary habits, coronary risk factors and mortality rates in Belgium. I Design and methodology. Acta Cardiol 1984;39:28592.[ISI][Medline]
23 Gambino SR, Freda VJ. The measurement of amniotic fluid bilirubin by the method of Jendrassik and Grof. Am J Clin Pathol 1966;46:198203.[ISI][Medline]
24 Fevery J, Muraca M, Mesa V, Van Steenbergen W, Blanckaert N. Plasma bilirubin pigments in health and disease. Molec Aspects Med 1987;9:391404.[ISI]
25 Villard PH, Herber R, Seree EM, Attolini L, Magdalou J, Lacarelle B. Effect of cigarette smoke on UDP-glucuronosyltransferase activity and cytochrome P450 content in liver, lung and kidney microsomes in mice. Pharmacol Toxicol 1998;82:7479.[ISI][Medline]
26 Subar AF, Harlan LC. Nutrient and food group intake by tobacco use status: The 1987 National Health Interview Survey. Ann NY Acad Sci 1993;686:31022.[ISI][Medline]
27 Bolton-Smith C, Woodward M, Brown CA, Tunstall-Pedoe H. Nutrient intake by duration of ex-smoking in the Scottish Heart Health Study. Br J Nutr 1993;69:31532.[ISI][Medline]
28 Vreman HJ, Wong RJ, Sanesi CA, Dennery PA, Stevenson DK. Simultaneous production of carbon monoxide and thiobarbturic acid reactive substances in rat tissue preparations by an iron-ascorbate system. Can J Physiol Pharmacol 1998;76:105765.[ISI][Medline]
29 Cross CE, O'Neil CA, Reznick AZ et al. Cigarette smoke oxidation of human plasma constituents. Ann NY Acad Sci 1993;686:7290.[Abstract]
30 Church DF, Pryor WA. Free-radical chemistry of cigarette smoke and its toxicological implications. Environ Health Perspect 1985;64:11126.[ISI][Medline]
31 Patrick DL, Cheadle A, Thompson DC, Diehr P, Koepsell T, Kinne S. The validity of self-reported smoking: a review and meta-analysis. Am J Public Health 1994;84:108693.[Abstract]
32 Wills TA, Cleary SD. The validity of self-reports of smoking: analyses by race/ethnicity in a school sample of urban adolescents. Am J Public Health 1997;87:5661.[Abstract]