Differences in body fat distribution and antioxidant status in Korean men with cardiovascular disease with or without diabetes

¹ From the Division of Cardiology, Yonsei Cardiovascular Center, Yonsei University, Seoul, Korea; the Department of Food and Nutrition, the College of Human Ecology, Yonsei University, Seoul, Korea; and the Nutrition Science and Services, Rehnborg Center for Nutrition and Wellness, the Nutrilite Division of Amway Corporation, Buena Park, CA.

² Supported in part by the Rehnborg Center for Nutrition and Wellness, Nutrilite Division of Amway Corporation, Buena Park, CA, and Amway Korea Ltd, Seoul, Korea.

³ Address reprint requests to JH Lee, Department of Food and Nutrition, College of Human Ecology, Yonsei University, Sudaemun-gu, Shinchon-dong 134, Seoul, Korea. E-mail: jhleeb{at}yonsei.ac.kr.

ABSTRACT  
Background: Abnormal body fat distribution and reduced antioxidant status have been shown to be effective markers of risk of cardiovascular disease (CVD).

Objective: The objective of this study was to determine the differences in body fat distribution and antioxidant status in healthy men (control subjects) and in men with CVD with or without diabetes.

Design: An oral-glucose-tolerance test was performed and CVD patients were subdivided into groups according to the presence or absence of diabetes. Adipose tissue areas were calculated from computed tomography scans made at the L1 and L4 vertebrae. Fasting serum concentrations of lipids, testosterone, insulin-like growth factor I, antioxidants, and plasma homocysteine were determined.

Results: There were no significant differences in mean age, body mass index (in kg/m²), or blood pressure between the groups. The visceral fat area at the L1 vertebra was nonsignificantly greater in CVD patients without diabetes than in control subjects, whereas it was significantly greater in CVD patients with diabetes than in control subjects at both the L1 and L4 vertebrae. Both groups of CVD patients had higher plasma concentrations of homocysteine and lower serum insulin-like growth factor I concentrations and superoxide dismutase activities than did control subjects. Serum ß-carotene and lycopene concentrations were lowest in the CVD patients with diabetes.

Conclusion: The concurrent presence of CVD and diabetes is associated with a greater negative effect on the risk factors typically associated with significant declines in health status.

Key Words: Visceral fat • antioxidants • diabetes • cardiovascular disease • Korean men

INTRODUCTION  
In addition to the traditional risk factors for cardiovascular disease (CVD) (hypertension, diabetes, smoking, and hypercholesterolemia), abdominal obesity, impaired antioxidant status, and hyperinsulinemia have been reported to be risk factors (1). An elevated fasting insulin concentration is associated with insulin resistance, which increases an individual's risk of CVD (2–4). Insulin resistance has also been reported to be closely associated with visceral fat accumulation (5). Therefore, central body fat distribution appears to be a greater risk factor for CVD than does obesity itself (6). Men with angiographically confirmed CVD had a higher waist-to-hip ratio (WHR) and a higher serum insulin concentration than did healthy men of similar ages and body mass indexes (BMI; in kg/m²) (7, 8).

The serum -tocopherol concentration was inversely correlated with WHR (r = - 0.23, P = 0.0001) in healthy Swedish adults (9). Carotenoids, glutathione peroxidase, superoxide dismutase, and -tocopherol are known to protect serum lipoproteins and vascular endothelium against oxidative stress and lipid peroxidation, which lead to atherosclerosis (10–12). Previous studies indicated that the concentrations of antioxidants in plasma and tissue were lower in patients with CVD than in a healthy control group and that the ß-carotene concentration in adipose tissue was significantly lower in patients with myocardial infarction than in the control group (13).

It is well accepted that a poor antioxidant status promotes free radical production and that such an increase in free radicals correlates with increases in total homocysteine (tHcy) concentrations. An elevated tHcy concentration is thought to be a risk factor for CVD and may accelerate the progression and the initial incidence of arteriosclerosis via direct cytotoxic effects on endothelial cells or via LDL oxidation due to free radical generation (14–16). Moderate hyperhomocysteinemia is defined as plasma concentrations of tHcy >15–16 µmol/L and has been identified in 20–40% of patients with atherosclerotic vascular disease (14, 17).

The risk of CVD is 2–3 times greater in men with diabetes than in men without diabetes (13). It was observed that risk factors such as hypertension, dyslipidemia, abdominal obesity, and insulin resistance occur frequently in persons with diabetes. In addition, the antioxidant defense system of such persons may be damaged by oxidative stress after glycosylation (13). CVD patients with abdominal obesity and insulin resistance may have a higher prevalence of type 2 diabetes mellitus.

An increase in morbidity and mortality rates as a result of CVD was observed in 1996 (18). However, little has been reported about the effects of central visceral fat accumulation, hyperinsulinemia, compromised antioxidant status, and elevated plasma tHcy on the incidence and progression of CVD in Korean men. We investigated whether there are differences in body fat distribution and antioxidant status among 3 groups: healthy men, CVD patients without diabetes, and CVD patients with diabetes.

SUBJECTS AND METHODS  
Subjects
Fifty-six men with CVD (aged 30–69 y) were referred by the Division of Cardiology, Yonsei Cardiovascular Center, Yonsei Severance Hospital, Seoul, Korea. The inclusion criteria required angiographic evidence with 50% occlusion of one or more major coronary arteries, previous myocardial infarction, or angina pectoris. The CVD patients were taking drugs such as blood pressure–lowering agents, aspirin, and antithrombosis agents, but none were taking drugs known to influence lipid concentrations. Just one diabetic patient was already taking an oral hypoglycemic agent.

Sixty-four healthy men were selected for the control group from volunteers who responded to advertisements for a nutrition study conducted by the Clinical Nutrition Research Team at Yonsei University in 1998. All control subjects had normal glucose tolerance and normal results on electrocardiograms and none had a diagnosis of any type of CVD (including coronary heart disease, stroke, and peripheral vascular disease) or cancer. All subjects gave their informed consent before participating in the study. After completing an oral-glucose-tolerance test, the CVD patients were subdivided into 2 groups: subjects with (n = 26) and without (n = 30) diabetes. The Institutional Review Board for Clinical Research of the Yonsei University (Severance Hospital) approved the study protocol and all subjects gave their informed consent.

Alcohol consumption, smoking status, dietary supplement intake, and anthropometry
Alcohol consumption was calculated as the grams of ethanol ingested per day and cigarette smoking data were reported as the number of cigarettes smoked per day. The body heights and weights of subjects in light clothing were measured and BMI was computed. Waist and hip circumferences were measured and the WHR was computed as an indication of the index of body fat distribution. A survey was used to determine the use of dietary supplements.

Computed tomography of the fat areas at the L1 and L4 vertebrae
To quantify fat area, all subjects underwent computed tomography (CT) scanning with a High Speed Advantage 9800 Scanner (General Electric, Milwaukee). Cross-sectional images were made of the abdomen at the L1 and L4 vertebrae. Each slice was analyzed for the cross-sectional area of fat by using a density contour program available with the CT scanner. The parameters for total abdominal fat density at the L1 and L4 vertebrae were between -150 and -50 Hounsfield units. Total abdominal fat area was divided into visceral and subcutaneous fat areas.

Blood collection and blood pressure measurement
Venous blood specimens were collected after a 12-h fast into plain (for serum glucose, lipid, antioxidant, and hormone analyses) or EDTA-treated (for tHcy analysis) tubes. The tubes for antioxidant analysis were immediately covered with aluminum foil and placed on ice in the dark until they arrived at the laboratory. Blood pressure was read from the left arm while subjects remained seated. An average of 3 measurements was recorded for each subject.

Serum lipids and lipoproteins
Serum total cholesterol, LDL cholesterol, and triacylglycerol were measured with commercially available kits on a Hitachi 7150 Autoanalyzer (Hitachi Ltd, Tokyo). After precipitation of serum chylomicrons, LDL, and VLDL with a precipitating agent, the HDL cholesterol left in the supernate was measured with an enzymatic method.

Oral-glucose-tolerance test
All subjects ingested a 75-g glucose solution after an overnight fast. Serum samples were collected before and 30, 60, 90, and 120 min after the glucose load. The criteria developed by the World Health Organization Expert Committee on Diabetes Mellitus was used to determine whether subjects had diabetes [a fasting glucose concentration >7.7 mmol/L (140 mg/dL) and a 2-h glucose concentration >11.1 mmol/L (200 mg/dL)]. Glucose was measured with a glucose oxidase method by using a glucose analyzer (Beckman Instruments, Irvine, CA). Insulin was measured by using an immunoradiometric assay (Dainabot, Tokyo) and C-peptide was measured by using a radioimmunoassay with commercial kits from Immuno Nucleo Corporation (Stillwater, MN). Each glucose, insulin, and C-peptide response was calculated as the area under the response curve.

Hormones
Serum testosterone was measured with a kit from Immuchem (Diagnostics Product Corp, Los Angeles). Serum insulin-like growth factor I (IGF-I) was measured with an immunoradiometric assay kit from Diagnostic System Laboratories (Webster, TX).

Serum sex hormone binding globulin and plasma total homocysteine
Radioimmunoassays were used to measure serum sex hormone binding globulin (SHBG). The free androgen index was the ratio of the percentage of testosterone to moles of SHBG (19). Plasma tHcy was analyzed with a Bio20 Autoloader Amino Acid Analyzer (Pharmacia Biotech, Cambridge, United Kingdom) with a postcolumn ninhydrin reaction system according to the modified method of Anderson et al (20) and Ueland et al (21).

Serum retinol, carotenoids, and tocopherols
Reversed-phase HPLC was used to determine retinol, carotenoids, and tocopherols in serum simultaneously as described by Yeum et al (22), Bankson et al (23), and McCrehan (24). The HPLC system consisted of an Alliance Waters 2690 separation module, a Waters 996 photodiode array detector, a Waters 474 scanning fluorescence detector, a C₁₈ Symmetry 3.9 x 15 cm column, and a PDA Millennium for version 2.15 DATA STATION (Waters, Milford, MA). The 474 scanning fluorescence detector was set at 294 nm for the detection of tocopherols. Retinol, carotenoids, and tocopherols were reported as serum concentrations uncorrected or corrected for the sum of serum cholesterol (mmol/L) and serum triacylglycerol (mmol/L) as suggested by Thurnham et al (25).

Serum antioxidant enzymes and malondialdehyde
Glutathione peroxidase activity was determined by using the coupled enzyme procedure at 340 nm, with hydrogen peroxide as substrate, according to Paglia and Valentine (26) and Deagen et al (27). Superoxide dismutase activity was estimated by the method of Marklund and Marklund (28) and Sheri et al (29). Malondialdehyde was assayed according to the fluorometric method described by Buckingham (30).

Statistical analysis
Statistical analyses were conducted with SPSS/PC⁺ (version 7.0; SPSS Inc, Chicago). For descriptive purposes, values are presented as means ± SEMs. To determine differences in body fat distribution and in antioxidant status between the 3 groups, the sample sizes were selected on the basis of the following assumptions: the present study would have 80% power to detect a 20% relative difference in tHcy concentrations between the healthy men and the 2 CVD patient groups ( = 0.05, two-tailed). Because these variables were reportedly 20% higher in a group of CVD patients than in a healthy group (31), differences between groups were evaluated by performing an analysis of variance followed by the Bonferroni test. A P value <0.05 was considered to indicate statistical significance.

RESULTS  
Alcohol consumption, cigarette smoking, and vitamin supplementation
Twenty-eight percent of the control subjects, 33% of the CVD patients without diabetes, and 15% of the CVD patients with diabetes were current smokers. Fifty-eight percent of the control subjects, 47% of the CVD patients without diabetes, and 38% of the CVD patients with diabetes typically consumed >5 g ethanol/d. Fifty-six percent of the control subjects, 30% of the CVD patients without diabetes, and 46% of the CVD patients with diabetes supplemented their diet with a multivitamin (B complex and vitamin C) preparation; however, none had taken any supplement for 1 mo before blood collection.

Age, BMI, blood pressure, and body fat distribution
There were no significant differences in age, BMI, or systolic or diastolic blood pressure between the 3 groups (Table 1). CVD patients with diabetes showed the highest mean WHR values. Visceral fat area at the L1 vertebra tended to be greater in the CVD patients without diabetes than in the control group. CVD patients with diabetes had a significantly greatest visceral fat area at the L1 and L4 vertebrae than did the control subjects (Figure 1).

View this table:
TABLE 1.. Age, anthropometric measures, and fat areas at different sites of the body in healthy men and in CVD patients with or without diabetes¹  

View larger version (17K):
FIGURE 1. . Mean (±SEM) visceral fat area and the ratio of visceral to subcutaneous fat area at the L1 and L4 vertebrae in healthy men (, control subjects; n = 64) and in CVD patients with (; n = 26) or without (; n = 30) diabetes. ²Significantly different from control subjects, P < 0.05 (ANOVA followed by the Bonferroni test).

 
Serum lipid, glucose, insulin, and C-peptide concentrations
Serum HDL-cholesterol concentrations were lowest in the CVD patients with diabetes (Table 2). Of the 3 groups, CVD patients with diabetes had the highest atherogenic index, fasting glucose concentration, and glucose response area. Response areas of insulin and C-peptide were significantly greater in the CVD patients without diabetes than in the control group.

View this table:
TABLE 2.. Serum lipids, glucose, insulin, and C-peptide and responses to a 75-g oral-glucose-tolerance test in healthy men and in CVD patients with or without diabetes¹  
SHBG, antioxidant enzyme activity, and plasma total homocysteine
The mean testosterone concentration was lower in the 2 groups of CVD patients than in the control group, but not significantly so (Table 3). Serum IGF-I concentrations were significantly lower in the 2 groups of CVD patients than in the control subjects. Serum SHBG, glutathione peroxidase, the free androgen index, and malondialdehyde were not significantly different between the 3 groups. Serum superoxide dismutase activities were significantly lower in the 2 groups of CVD patients than in the control subjects. tHcy concentrations were significantly higher in the 2 groups of CVD patients than in the control group (Figure 2).

View this table:
TABLE 3.. Sex hormones and antioxidant enzyme activities in healthy men and in CVD patients with or without diabetes¹  

View larger version (10K):
FIGURE 2. . Mean (±SEM) plasma homocysteine concentrations and serum total superoxide dismutase (SOD) activity in healthy men (, control subjects; n = 64) and in CVD patients with (; n = 26) or without (; n = 30) diabetes. ²Significantly different from control subjects, P < 0.05 (ANOVA followed by the Bonferroni test).

 
Serum retinol, carotenoids, and tocopherols
There were no significant differences in uncorrected and lipid-corrected serum concentrations of -carotene, retinol, and - or -tocopherol between the 3 groups (Table 4). The corrected and uncorrected serum cryptoxanthin concentrations were lower in the CVD patients with diabetes than in the control subjects, but not significantly so. The CVD patients with diabetes had the lowest uncorrected and lipid-corrected lycopene concentrations, and uncorrected ß-carotene concentrations in this group were lower (NS) than in the control group (Figure 3).

View this table:
TABLE 4.. Serum concentrations of carotenoids, retinol, and tocopherol in healthy men and in CVD pateints with or without diabetes¹  

View larger version (10K):
FIGURE 3. . Mean (±SEM) serum ß-carotene and lycopene concentrations in healthy men (, control subjects; n = 64) and in CVD patients with (; n = 26) or without (; n = 30) diabetes. ²Significantly different from control subjects, P < 0.05 (ANOVA followed by the Bonferroni test).

 

DISCUSSION  
The results indicate that visceral fat accumulation is higher in CVD patients than in healthy men of similar ages and BMIs and that excessive visceral fat accumulation may accelerate the progression of type 2 diabetes mellitus. Visceral fat accumulation induces insulin resistance, which encourages pancreatic ß-cells to secrete more insulin to maintain normal glucose metabolism. Thus, hyperinsulinemia results, which exacerbates insulin resistance and further increases the accumulation of upper body fat (4). Excessive accumulation of visceral fat may result in a relative deficiency of insulin, the secretion of which is insufficient to regulate insulin resistance and leads to impaired glucose tolerance and potentially to type 2 diabetes mellitus (32, 33).

Compared with the healthy control group with normal glucose tolerance, the CVD patients with diabetes showed severe insulin resistance, resulting in hyperglycemia when insulin concentrations were not maintained. The similar response areas of C-peptide and insulin in the control subjects and the CVD patients with diabetes indicate that further visceral fat accumulation and a relative deficiency of insulin may lead to a greater incidence of type 2 diabetes mellitus in CVD patients. This suggests that further visceral fat accumulation and a relative deficiency of insulin in those with CVD may lead to a greater incidence of type 2 diabetes mellitus.

Insulin resistance has been reported to influence lipoprotein lipase activity, lipoprotein metabolism, and glucose metabolism (4). Decreased lipoprotein lipase activity due to insulin resistance may delay triacylglycerol-rich lipoprotein uptake and decrease HDL-cholesterol formation. Another study reported that insulin resistance accelerated HDL-cholesterol catabolism, which led to a decrease in serum HDL-cholesterol concentrations (34). In addition to insulin resistance, serum testosterone concentrations in men were positively related to serum HDL cholesterol and were presumably related to the regulation of hepatic lipase activity (35).

An elevated serum testosterone concentration in men increases the number of ß-receptors in visceral fat cells and the activity of hormone-sensitive lipase, leading to fat mobilization (36). Growth hormone activity is also more apparent in visceral than in subcutaneous fat and it accelerates the lipolysis of visceral fat in the presence of testosterone. In the present study, IGF-I and testosterone concentrations were significantly different between the control subjects and the CVD patients with and without diabetes. These differences in hormone concentrations may partly reflect the difference in visceral fat area between the CVD patients with or without diabetes.

The CVD patients without diabetes had moderate hyperhomocysteinemia (mean tHcy concentration: 15.1 µmol/L), whereas the CVD patients with diabetes had mild hyperhomocysteinemia (mean tHcy concentration: 13.1 µmol/L). It has been suggested that tHcy may have direct cytotoxic effects or produce free radicals that can result in endothelial damage and dysfunction (14). An increase in free radical production contributes to oxidative stress and exhausts the antioxidant system. Therefore, a high tHcy concentration may be associated with low antioxidant capacity (12, 14). In the present study, hyperhomocysteinemic CVD patients with or without diabetes had lower superoxide dismutase activity than did the control subjects. Superoxide dismutase, a major antioxidant enzyme, catalyzes the dismutation of superoxide anions to hydrogen peroxide and oxygen. The antioxidant defense system includes enzymatic and nonenzymatic antioxidants. The most prominent nonenzymatic antioxidants are vitamins E and C and carotenoids in the cytoplasm and lipid membranes of cells (37).

ß-Carotene is a fat-soluble carotenoid and antioxidant that is well known for its relation to atherosclerosis. ß-Carotene is concentrated in lipoprotein fractions such as HDL and LDL and accumulates in atherosclerotic plaques (38, 39). It is well established that high serum ß-carotene concentrations may protect against endothelial damage by decreasing cholesterol uptake into those cells and that ß-carotene inhibits atherosclerotic lesion formation (39, 40).

In the present study, serum ß-carotene concentrations were associated with physiologic and lifestyle factors, including carotenoid intake, sex, age, BMI, and smoking status (41, 42). The lower serum ß-carotene concentration in CVD patients without diabetes than in the control group may have been related not only to CVD itself, but also to this group's lower carotenoid intakes and current smoking status. The lower serum concentrations of ß-carotene and lycopene in the CVD patients with diabetes than in the other 2 groups may have resulted from the increased oxidative stress due to diabetes rather than to behavioral or dietary choices. Persons with diabetes appear to be more susceptible to ambient oxidative stress and might be under greater oxidative stress than are persons without diabetes because of glycosylation followed by oxidation processes (13).

Oxidative stress affects the peroxidation of polyunsaturated fatty acids in LDL and exhausts ubiquinone-10 as well as vitamin E and carotenoids, including lycopene, cryptoxanthin, and lutein (44). Each of these compounds, singly, contributes little to the resistance to LDL oxidation, but collectively may contribute a great deal. Therefore, ingestion of a combination of different antioxidants rather than of a single micronutrient may have a better effect against LDL oxidation (43).

In summary, the male CVD patients in the present study had greater visceral fat accumulation and insulin and tHcy concentrations and lower IGF-I, superoxide dismutase, and ß-carotene concentrations than did the control subjects, who were healthy men matched for age and BMI. In addition, the CVD patients with diabetes had the greatest visceral fat areas and tHcy concentrations and the lowest serum concentrations of HDL cholesterol and carotenoids of the 3 groups. Given the significant differences observed between the 3 groups, lifestyle modifications such as smoking cessation, reductions in abdominal fat, and consumption of antioxidant-rich foods are recommended to delay the progression of CVD and to maintain a healthy life. In addition, given the significantly lower serum antioxidant concentrations in the CVD patients than in the control subjects, it may be prudent for patients with CVD to supplement their diet with antioxidants such as ubiquinone-10, vitamins E and C, and mixed carotenoids, including ß-carotene, lycopene, cryptoxanthin, and lutein.

REFERENCES  

1. Hercberg S, Galan P, Preziosi P, Alfarez MJ, Vazquez C. The potential role of antioxidant vitamins in preventing cardiovascular diseases and cancers. Nutr 1998;14:513–20.
2. Huh KB, Lee HC, Cho SY, Lee JH, Song YD. The role of insulin resistance in Korean patients with coronary atherosclerosis. Diabetes 1996;45(suppl):S59–61.
3. Colman E, Toth MJ, Katzel LI, Fonong T, Gardner AW, Poehlman ET. Body fatness and waist circumference are independent predictors of the age-associated increase in fasting insulin concentrations in healthy men and women. Int J Obes Relat Metab Disord 1995;19: 798–803.
4. Despres J-P, Marette A. Relation of components of insulin resistance syndrome to coronary disease risk. Curr Opin Lipidol 1994;5:274–89.
5. Cefalu WT, Wang ZQ, Werbel S, et al. Contribution of visceral fat mass to the insulin resistance of aging. Metabolism 1995;44:954–9.
6. Zamboni M, Armellini F, Cominacini L, et al. Obesity and regional body-fat distribution in men: separate and joint relationships to glucose tolerance and plasma lipoproteins. Am J Clin Nutr 1994; 60:682–7.
7. Cha BS, Song YD, Paik IK, et al. Hyperinsulinemia in patients with coronary heart disease. J Korean Diabetes Assoc 1996;20:291–302.
8. Hauner H, Bognar E, Blum A. Body fat distribution and its association with metabolic and hormonal risk factors in women with angiographically assessed coronary artery disease. Evidence for the presence of a metabolic syndrome. Atherosclerosis 1994;105:209–16.
9. Orvall M, Tengblad S, Bessby B. Lower tocopherol serum concentrations in subjects with abdominal obesity. J Intern Med 1993; 234:53–60.
10. Clinton SK. Lycopene: chemistry, biology, and implications for human health and disease. Nutr Rev 1998;56:35–51.
11. Bonithon-Kopp C, Coudray C, Berr C, et al. Combined effects of lipid peroxidation and antioxidant status on carotid atherosclerosis in a population aged 59–71 years: The EVA Study. Am J Clin Nutr 1997;65:121–7.
12. Buczynski A, Wachowicz B, Kedziora-Kornatowska K, Tkaczewski W, Kedziora J. Changes in antioxidant enzyme activities, aggregability and malondialdehyde concentration in blood platelets from patients with coronary heart disease. Atherosclerosis 1993;100:223–8.
13. Kardinaal AFM, Kok FJ, Ringstad J, et al. Antioxidants in adipose tissue and risk of myocardial infarction: the EURAMIC Study. Lancet 1993;342:1379–84.
14. Currie IC, Wilson YG, Scott J, et al. Homocysteine: an independent risk factor for the failure of vascular intervention. Br J Surg 1996; 83:1238–41.
15. Woodside JV, Yarnell JWG, McMaster D, et al. Effect of B-group vitamins and antioxidant vitamins on hyperhomocysteinemia: a double-blind, randomized, factorial-design, controlled trial. Am J Clin Nutr 1998;67:858–66.
16. Pancharumiti N, Lewis C, Sanverlich HE, et al. Plasma homocyst(e)ine, folate, and vitamin B-12 concentrations and risk for early-onset coronary artery disease. Am J Clin Nutr 1994;59:940–8.
17. Duell PB, Malinow MR. Homocysteine: an important risk factor for atherosclerotic vascular disease. Curr Opin Lipidol 1998;8:28–34.
18. National Statistical Office. Annual report on the cause of death statistics—based on vital registration. Republic of Korea: National Statistical Office, 1996.
19. Carter GD, Holland SM, Alaghband-Zadeh J, Dorrington-Ward P, Wise PH. Investigation of hirsutism: testosterone is not enough. Ann Clin Biochem 1983;20:262–3.
20. Andersson A, Battstrom L, Isaksson A, Israelsson B, Hultberg B. Determination of homocysteine in plasma by ion-exchange chromatography. Scand J Clin Lab Invest 1989;49:445–9.
21. Ueland PM, Refsum H, Stabler SP, Mailnow MR, Andersson A, Allen RH. Total homocysteine in plasma or serum: method and clinical applications. Clin Chem 1993;39:1764–79.
22. Yeum K-J, Lee-Kim YC, Yoon S, et al. Similar metabolites formed from ß-carotene by human gastric mucosal homogenates, lipoxygenase, or linoleic acid hydroperoxide. Arch Biochem Biophys 1995;321:167–74.
23. Bankson DD, Russell RM, Sadowski JA. Determination of retinyl esters and retinol in serum or plasma by normal phase liquid chromatography, method and applications. Clin Chem 1986;32:35–40.
24. McCrehan WA. Determination of retinol, -tocopherol and ß-carotene in serum by liquid chromatography. Methods Enzymol 1986;189: 172–81.
25. Thurnham DI, Davis JA, Crump BJ, Situnayake RD, Davis M. The use of different lipids to express serum tocopherol: lipid ratios for the measurement of vitamin E status. Ann Clin Biochem 1986;23:514–20.
26. Paglia DE, Valentine WN. Studies on the quantitative and qualitative characterization of erythrocyte glutathione peroxidation. J Lab Clin Med 1967;70:158–69.
27. Deagen JT, Butler JA, Beilstein MA, Whanger PD. Effects of dietary selenite, selenocystine and selenomethionine on selenocysteine lyase and glutathione peroxidase activities and on selenium levels in rat tissues. J Nutr 1987;117:91–8.
28. Marklund S, Marklund G. Involvement of the superoxide anion radical in the autoxidation of pyrogallol and a convenient assay for superoxide dismutase. Eur J Biochem 1974;47:469–74.
29. Sheri ZC, Keen CL, Hurley LS. SOD activity and lipid peroxidation in the rat: developmental correlations affected by manganese deficiency. J Nutr 1983;113:2498–504.
30. Buckingham KW. Effect of dietary polyunsaturated/saturated fatty acid ratio and dietary vitamin E on lipid peroxidation in the rat. J Nutr 1985;115:1425–35.
31. Genest JJ Jr, McNamara JR, Salem DN, Wilson PWF, Schaefer EJ, Malinow MR. Plamsa homocyst(e)ine levels in men with premature coronary artery disease. J Am Coll Cardiol 1990;16:1114–9.
32. Frayn KN. Insulin resistance and lipid metabolism. Curr Opin Lipidol 1993;4:197–204.
33. Reaven GM. The role of insulin resistance and hyperinsulinemia in coronary heart disease. Metabolism 1992;41:16–26.
34. Semmens J, Rouse I, Belin LJ, Masarei RL. Relationship of plasma HDL-cholesterol to testosterone, estradiol and sex hormone-binding globulins in men and women. Metabolism 1983;32:429–32.
35. Lichtenstein MJ, Yarnell JWG, Elwood PC, et al. Sex hormones, insulin, lipids and prevalent ischemic heart disease. Am J Epidemiol 1987;126:647–57.
36. Bjorntorp P. The regulation of adipose tissue distribution in humans. Int J Obes Relat Metab Disord 1996;20:291–302.
37. Allard JP, Aghdassi E, Chau J, Salit I, Walmsley S. Oxidative stress and plasma antioxidant micronutrients in humans with HIV infection. Am J Clin Nutr 1998;67:143–7.
38. Stahl W, Sies H. Antioxidant defense: vitamin E and C and carotenoids. Diabetes 1997;46:S14–8.
39. Buring JE, Hennekens CH. Antioxidant vitamins and cardiovascular disease. Nutr Rev 1997;55:S53–60.
40. Vogel S, Contois JH, Tucker KL, Wilson PWF, Schafer EJ, Lammi-Keefe CJ. Plasma retinol and plasma and lipoprotein tocopherol and carotenoid concentrations in healthy elderly participants of the Framingham Heart Study. Am J Clin Nutr 1997;66:950–8.
41. Kaplan LA, Stein EA, Willett WC, Stampfer MJ, Stryker WS. Reference range of retinol, tocopherols, lycopene and alpha- and beta-carotene in plasma by simultaneous high-performance liquid chromatographic analysis. Clin Physiol Biochem 1987;5:297–304.
42. Halliwell B. Antioxidants and human disease: a general introduction. Nutr Rev 1997;55:S44–52.
43. Reunanen A, Knekt P, Aaran RK, Aromaa A. Serum antioxidants and risk of non-insulin dependent diabetes mellitus. Eur J Clin Nutr 1998;52:89–93.