To explore the independent associations of body height, body mass index (BMI), waist circumference and hip circumference with high-density lipoprotein-cholesterol (HDL-cholesterol) and non-high-density lipoprotein-cholesterol (non-HDL-cholesterol), in a large general population sample.
Urban and rural areas throughout Greece.
In total,10 837 volunteers, 2034 men and 8803 women, aged 25–82 years, participants in the European Prospective Investigation into Cancer and Nutrition study (EPIC), who have never smoked and never been treated for dyslipidemia.
The effect of height on non-HDL-cholesterol was opposite but in absolute terms almost as important as that of BMI with no gender interaction. Among women, hip circumference was inversely associated with non-HDL-cholesterol (standardized coefficient bst=−1.11, with standard error (s.e.)=0.42) and positively with HDL-cholesterol (bst=0.85, s.e.=0.12) whereas, waist circumference was inversely associated with HDL-cholesterol (bst=−1.16, s.e.=0.13) and strongly positively with non-HDL-cholesterol (bst=8.83, s.e.=0.45). Among men, associations were generally weaker (in absolute terms by about 50%) and for hip circumference the association with non-HDL-cholesterol was actually non significantly positive.
Height was inversely associated with HDL and non-HDL-cholesterol implicating early life phenomena in the regulation of these variables. Larger hip circumference among women had beneficial effects on blood cholesterol fractions by increasing HDL-cholesterol and reducing non-HDL-cholesterol, whereas among men the relevant effects were less clear cut. The detrimental consequences of large waist circumference on both HDL (reduction) and non-HDL-cholesterol (increase) were also particularly marked among women.
The European Prospective Investigation into Cancer and Nutrition (EPIC) is coordinated by the International Agency for Research on Cancer (World Health Organization) and supported by the Europe Against Cancer Program of the European Commission. The Greek segment of the EPIC study is also supported by the Greek Ministry of Health and the Greek Ministry of Education. This study was additionally supported by the fellowship ‘Vassilios and Nafsika Tricha’.
It has long been known that body mass index (BMI) and waist circumference, frequently assessed as waist to hip ratio, are important risk factors for cardiovascular and other chronic diseases (National Task Force on the Prevention and Treatment of Obesity, 2000; Despres et al., 2001; Aronne & Segal, 2002; Pi-Sunyer, 2002; Kanaya et al., 2003). Hip circumference has also attracted the interest of investigators as a possible independent risk factor for cardiovascular diseases, with evidence accumulating that it is inversely associated with morbidity and mortality from these diseases (Hunter et al., 1997; Williams et al., 1997; Lissner et al., 2001; Seidell et al., 2001; Snijder et al., 2004b; Sakai et al., 2005). Furthermore, the interest on metabolic syndrome highlighted the role of adipose tissue distribution in human pathology (Carr et al., 2004; Wahrenberg et al., 2005). Although height has been inversely associated with the incidence of coronary heart disease (Peck and Vagero, 1989; Palmer et al., 1990; Yarnell et al., 1992; Hebert et al., 1993; Kannam et al., 1994; Krahn et al., 1994; Parker et al., 1998; Wamala et al., 1999; Forsen et al., 2000), few investigations have examined whether this variable is associated with one or more aspects of the metabolic syndrome, over and beyond its presence in the denominator of BMI (Henriksson et al., 2001; Gunnell et al., 2003; La Batide-Alanore et al., 2003; Lawlor et al., 2004). The study of the risk implications of anthropometric variables, notably height, weight, waist circumference and hip circumference, has been hindered by the fact that these variables are strongly interrelated. This introduces collinearity problems in regression analyses, as well as, interpretation problems, since inferences on associations of anthropometric variables with health outcomes are mutually conditional. A procedure that bypasses the problem is the use of the residuals method (Willett and Stampfer, 1986) that generates largely uncorrelated anthropometric variables, the risk implications of which can be directly assessed. Such a procedure has been previously applied by Seidell et al. (2001) who have documented independent and opposite effects on cardiovascular risk factors of waist and hip circumferences in a study of over 600 subjects. We have exploited the availability of total and high density lipoprotein (HDL) blood cholesterol measurements for over 10 000 adult men and women who have never smoked, in order to evaluate the independent associations of HDL and non-HDL-cholesterol with height, BMI, waist circumference and hip circumference.
Subjects and methods
The study sample initially consisted of 11 856 participants, 2166 men and 9690 women, aged 25–82 years, never-smokers, who voluntarily enrolled in the Greek EPIC study during 1994–1999, and for whom total cholesterol and high density lipoprotein cholesterol (HDL-cholesterol) measurements were available (Benetou et al., 2000). EPIC (European Prospective Investigation into Cancer and nutrition) is a multicountry, prospective, cohort study, coordinated by the International Agency for Research on Cancer (IARC), investigating the role of nutrition and other lifestyle and environmental factors in the etiology of cancer and other chronic diseases (Riboli and Kaaks, 1997; Riboli et al., 2002). The large preponderance of women reflects, partly the intentional oversampling of women in the Greek EPIC study (since one of the primary objectives was to explore the nutritional etiology of breast cancer) and partly the very high prevalence of smokers among men in Greece.
One hundred and sixteen subjects with unreasonable values for BMI (>70 kg/m2) or energy intake (<750 kcal) were excluded. Also, excluded were 896 (7.6%) subjects who have reported to systematically receive, or have received, pharmaceutical treatment for dyslipidemia, in order to avoid any possible interference of the effect of such treatment with possible associations of height, BMI, waist circumference and hip circumference with HDL-cholesterol and non-HDL-cholesterol. Another seven study participants were excluded because of missing values. Therefore, the final study sample included 10 837 participants, 2034 (19%) men and 8803 (81%) women, who have never smoked and have never received drug treatment for dyslipidemia. All procedures were in line with the Helsinki declaration for human rights, all volunteers signed informed consent forms, and the study protocol was approved by ethical committees at IARC and at the University of Athens Medical School.
After signing the informed consent form, participants completed an extensive interviewer-administered questionnaire with information on various socio-demographic variables, medical history and lifestyle characteristics. In addition, each EPIC participant underwent a baseline clinical examination during which anthropometric (weight, height, as well as waist and hip circumferences) measurements were performed, blood pressure was measured and a nonfasting blood sample was collected.
Concerning anthropometry, height and weight were measured while subjects were standing without shoes and lightly clothed. Weight was measured to the nearest 100 g with regularly calibrated digital scales and height was measured to the nearest 0.1 cm. Waist and hip measurements were undertaken using an inelastic tape without compressing the skin and were recorded to the nearest 0.1 cm, while participants wore no restrictive underwear. Measurements were taken at the end of the normal respiration with the participant standing erect, with the arms at the side and the feet together. Waist circumference was measured around the smallest circumference between the lowest rib and iliac crest, or, for obese subjects with no natural waist, midway between the lowest rib and iliac crest. Hip circumference was measured horizontally at the level of the greatest lateral extension of the hips. BMI was calculated as the ratio of weight in kg divided by the square of the height in meters. This metric was developed to allow evaluation of degree of obesity independently of height (Billewicz et al., 1962; Khosla and Lowe, 1967).
One ml of plasma from each participant was used for the laboratory analysis. Total blood cholesterol was determined by the enzymatic method CHOD-PAP and HDL-cholesterol by the phosphotungstic method (Benetou et al., 2000). Non-HDL cholesterol values were calculated by subtracting HDL-cholesterol from total cholesterol values. The collaborating laboratory (BIO-CHECK, Athens) was a member of the National System of External Quality Control for Clinical Laboratories. The intrasample precision coefficient of variation (CV) for the measurement of total and HDL blood cholesterol was in each case less than 3%. Plasma values were converted into serum values with the use of a curve created from the determination of total cholesterol and HDL-cholesterol in plasma/serum pairs.
All statistical analyses were carried out separately for each gender. Descriptive statistical analysis was performed with simple tabulations. In the main analysis, the data were modeled through multiple linear regression with non-HDL-cholesterol and HDL cholesterol, alternatively, as the dependent variables. All the analyses were performed with the STATA statistical package (Intercooled Stata 7.0) and a level of 5% was used to denote statistical significance.
The focus of this study was the assessment of independent associations of height, weight, waist circumference and hip circumference with HDL-cholesterol and non-HDL-cholesterol. Introduction of all four variables into the same model accommodated mutual confounding but it is conceptually awkward because, for example, height is specified for a person with given weight, waist and hip circumference. It is easy to bypass this problem by introducing BMI instead of weight (height and BMI are uncorrelated). In order to accomplish the same objective with respect to waist circumference and hip circumference we have used a process that is well known in statistics (the residuals method), has been introduced in nutritional epidemiology by Willett and Stampfer (1986) and has been previously utilized by Seidell et al. (2001) in the evaluation of independent associations of waist and hip circumferences with cardiovascular risk factors. In order to apply this method in our data we have regressed, alternatively, waist circumference and hip circumference, separately for men and women, on height and BMI, and we have calculated for each participant waist circumference and hip circumference the corresponding residuals from the respective regression lines (waist residual and hip residual).
Independent variables in final regression models were height, BMI, waist circumference residual and hip circumference residual with their associations with HDL-cholesterol and non-HDL-cholesterol expressed per one standard deviation increment to facilitate comparability and interpretability. Also included in the models were age (categorically: less than 45, 45–54, 55–64, 65 or more years) and years of education as an indicator of socioeconomic status (categorically: 0–5 years, 6 years, 7–12 years, more than 12 years).
In Table 1 the distribution of study participants by gender and age at enrollment is shown. There are by design more women than men in the cohort. Very few participants were younger than 35 years and, when age was controlled for in the analyses, these were included in the less than 45 years age group.
Table 2 shows mean values and standard deviations of the principal study variables, among men and women. The data in this middle-aged general population sample reveal the expected patterns, in that men, in comparison to women, are taller, somewhat less heavy, have larger waist circumference and smaller hip circumference. Total cholesterol is slightly higher in women mostly because women have high levels of HDL-cholesterol.
The correlation matrices between the anthropometric variables, before and after transformation, among men and women, are depicted in Table 3. It is evident that the transformation minimizes correlation coefficients among the study variables.
Table 4 shows multiple regression derived, adjusted for age and educational level, as well as mutually among the indicated variables, standardized partial regression coefficients of non-HDL-cholesterol and HDL-cholesterol on height and BMI (Model 1), or on height, BMI, hip circumference residual and waist circumference residual (Model 2). Distinct models are used for men and women. Standardized partial regression coefficients indicate how much the dependent variable (HDL or non-HDL-cholesterol) changes when each of the independent (predictor) variables increases by one standard deviation while the other independent variables are kept constant.
When hip and waist circumferences are not included in the models (Model 1), BMI is significantly positively associated with non-HDL cholesterol among both men and women. The respective standardized regression coefficients are mutually compatible (nonsignificant interaction with gender, P-value=0.79). Interestingly, however, there are significant inverse associations, in both genders, of height with non-HDL-cholesterol and the respective standardized regression coefficients are, in absolute terms, half as large as the corresponding standardized regression coefficients for BMI. When hip and waist circumference are included in the models in addition to height and BMI (Model 2) the absolute values of the standardized regression coefficients for height considerably increase becoming similar to those for BMI. Waist circumference, after adjustment for height and BMI through the residuals method, as well as for hip circumference, is significantly positively associated with non-HDL-cholesterol and the respective standardized regression coefficient is considerably higher among women than among men (statistically significant interaction with gender, P-value <0.001). More importantly, whereas hip circumference is positively associated with non-HDL-cholesterol among men (P-value=0.098), it is significantly inversely associated among women (statistically significant interaction with gender, P-value=0.002).
With respect to HDL-cholesterol, when height and BMI, but not waist and hip circumference, are included in the models, significant inverse associations are evident, with standardized regression coefficients being, in absolute terms, four times as large for BMI as for height (Model 1). When hip and waist circumference are included in the models in addition to height and BMI (Model 2) the standardized regression coefficients for height and BMI change very little among both men and women. Waist circumference is inversely associated with HDL-cholesterol in both genders, but the association is stronger among women (statistically significant interaction with gender, P-value=0.002). In contrast, hip circumference is positively associated with HDL-cholesterol in both genders, although the association is statistically significant only among women and there is suggestive evidence of gender interaction (P-value=0.09).
Owing to the relatively large age spectrum of the study participants, we have explored whether there might be interactions of age with the anthropometric variables under consideration with respect to HDL and non-HDL-cholesterol. The interaction terms were evaluated for age groups less than 50 years, 50–64 years and 65 or more years, putting more emphasis on biological cutoff points rather than statistical efficiency. Many interaction terms were evaluated so that statistical significance should be interpreted with caution. Among women, there is a significant interaction of age by height with respect to non-HDL-cholesterol (P-value ∼ 0.001) in that the regression coefficient is strongly negative among women less than 50 years old (−3.46) and 50–64 years old (−3.37), but only weakly negative among women older than 65 years old (−0.99). Nominally significant interactions of age with height with respect to non-HDL cholesterol among men (P-value ∼0.023) and of age with BMI with respect to non-HDL cholesterol (P-value <0.001) as well as HDL-cholesterol (P-value ∼0.015) among women are not consistent. Thus, the age-specific regression coefficients of the anthropometric variables with the blood cholesterol fractions do not show monotonic trends with age, that is the regression coefficients first decrease and then increase, or vice versa, with increasing age.
In a study of over 10 000 adult men and women, who have never smoked and have never been treated for dyslipidemia, we have confirmed that BMI is positively associated with non-HDL-cholesterol and inversely associated with HDL-cholesterol in both genders. We have also observed, however, that height is inversely associated with non-HDL-cholesterol and, to a lesser extent, HDL-cholesterol in both genders. Thus, not only heavier individuals, but also shorter individuals, tend to have less favourable levels of cholesterol fractions. Large waist circumference is detrimental for both men and women, but more so for women than for men. Lastly, hip circumference is essentially neutral for men because it increases both HDL and non-HDL-cholesterol to a proportionally similar degree, but it is highly protective for women because it considerably increases HDL-cholesterol and reduces non-HDL-cholesterol. Differential modulation of free fatty acid levels or adipokines by adipose tissue distribution patterns is a reasonable and testable hypothesis (Frayn, 2002; Ravussin and Smith, 2002; Snijder et al., 2004a).
The positive association of excess body fat, specifically abdominal fat, with dyslipidemia (defined as high LDL-cholesterol, low HDL-cholesterol as well as high VLDL cholesterol and triglycerides) has been firmly established by numerous epidemiologic studies conducted in various geographical regions, both genders and variable ethnic groups (Denke et al., 1993, 1994; Despres, 1994; Brown et al., 2000; Hu et al., 2000a, 2000b; Thomas et al., 2004). The association has been further supported by a plausible pathophysiologic mechanism of action which mainly incriminates the high amounts of free fatty acids levels, observed in obesity (Sheehan and Jensen, 2000; Aronne and Segal, 2002; Pi-Sunyer, 2002).
The inverse association of height with non-HDL-cholesterol has also been reported by other investigators (Henriksson et al., 2001; La Batide-Alanore et al., 2003) and is likely to be real. The clinical consequence of this association is probably small but the phenomenon is intriguing. In contrast to BMI and body fat distribution which change through life and have a plausible latency in their connection with adult life blood lipids, height is determined during the first 20 years of life and a link with blood lipids four decades later requires the invocation of a novel physiologic mechanism. In a possibly relevant analogy, it has been reported that postnatal cholesterol intake affects blood cholesterol homeostasis in later life. Specifically, a relatively high cholesterol intake during the immediate postnatal period has been associated with more effective catabolism of the compound in adult life and lower blood cholesterol levels (Marmot et al., 1980; Innis, 1985; Kolacek et al., 1993; Cruz et al., 1994; Bergstrom et al., 1995). Thus, we can speculate that growth in childhood, adolescence and early adulthood may somehow affect the setting of parameters of cholesterol metabolism throughout life, for example, by means of longitudinal ‘tracking’ of high or low levels of growth hormone that is inversely associated with blood cholesterol (Abdu et al., 2001; Cianfarani et al., 2002; Takada et al., 2003). The fact that the inverse association of height with non-HDL-cholesterol weakens with age (more evidently in women) may be explained by the gradual decline of height with progressing age in many subjects (Rossman, 1977; Svanborg and Mellstrom 1991) and the concomitantly increasing nondifferential misclassification.
Non-HDL-cholesterol (the sum of low density plus very-low-density lipoprotein) has been recognised as an acceptable surrogate marker of apolipoprotein B, the major apolipoprotein of all atherogenic lipoproteins (National Cholesterol Education Panel III, 2002). Consistent with this relation, studies have shown that non-HDL-cholesterol is at least as good or even better predictor of cardiovascular disease risk in comparison to LDL-cholesterol and apolipoproteins A-1 and B. (Cui et al., 2001; Shai et al., 2004; Ridker et al., 2005). In the absence of data on triglyceride levels, non-HDL-cholesterol is a good indicator of atherogenic lipoproteins in people with high triglycerides (200–499 mg/dl) while simultaneously correlates well with LDL-cholesterol in persons with normal triglycerides (National Cholesterol Education Panel III, 2002). Furthermore, non-HDL-cholesterol does not require a fasting state and has been proposed as a screening tool for cardiovascular risk (Farwell et al., 2005).
Strengths of this investigation are the large sample size, restriction of the study to never smokers and individuals who have never been treated for dyslipidemia and use of an analysis strategy that generates internally comparable and directly interpretable results. Limitations are that blood lipids other than total and HDL-cholesterol have not been determined, and possible weak confounding by factors such as physical activity and energy intake (over and beyond their balance reflected in BMI) have not been addressed.
In conclusion, in a large, general population-based, study we have found evidence that height is inversely associated with HDL and non-HDL-cholesterol implicating early life phenomena in the regulation of important cardiovascular risk factors. Furthermore, our results indicate that larger hip circumference among women, conditionally on height, BMI and waist circumference has beneficial effects on blood cholesterol fractions by increasing HDL-cholesterol and reducing non-HDL-cholesterol in the blood. Among men the relevant effects are less clear cut (hip circumference is positively associated with both HDL and non-HDL-cholesterol). Finally, our findings document that the detrimental consequences of large waist circumference on both HDL (reduction) and non-HDL-cholesterol (increase) are particularly marked among women.
Abdu TA, Neary R, Elhadd TA, Akber M, Clayton RN (2001). Coronary risk in growth hormone deficient hypopituitary adults: increased predicted risk is due largely to lipid profile abnormalities. Clin Endocrinol (Oxford) 55, 209–216.
Aronne LJ, Segal KR (2002). Adiposity and fat distribution outcome measures: assessment and clinical implications. Obes Res 10, 14S–21S.
Benetou V, Chloptsios Y, Zavitsanos X, Karalis D, Naska A, Trichopoulou A (2000). Total cholesterol and HDL-cholesterol in relation to socioeconomic status in a sample of 11, 645 Greek adults: the EPIC study in Greece. Scand J Public Health 28, 260–265.
Bergstrom E, Hernell O, Persson LA, Vessby B (1995). Serum lipid values in adolescents are related to family history, infant feeding, and physical growth. Atherosclerosis 117, 1–13.
Billewicz WZ, Thomson AM, Kemsley WFF (1962). Indices of adiposity. Br J Prev Soc Med 16, 183–188.
Brown CD, Higgins M, Donato KA, Rohde FC, Garrison R, Obarzanek E et al. (2000). Body mass index and the prevalence of hypertension and dyslipidemia. Obes Res 8, 605–619.
Carr DB, Utzschneider KM, Hull RL, Kodama K, Retzlaff BM, Brunzell JD et al. (2004). Intra-abdominal Fat is a major determinant of the National Cholesterol Education Program Adult Treatment Panel III Criteria for the metabolic syndrome. Diabetes 53, 2087–2094.
Cianfarani S, Geremia C, Scott CD, Germani D (2002). Growth, IGF system, and cortisol in children with intrauterine growth retardation: is catch-up growth affected by reprogramming of the hypothalamic-pituitary-adrenal axis? Pediatr Res 51, 94–99.
Cruz ML, Wong WW, Mimouni F, Hachey DL, Setchell KD, Klein PD et al. (1994). Effects of infant nutrition on cholesterol synthesis rates. Pediatr Res 35, 135–140.
Cui Y, Blumenthal RS, Flaws JA, Whiteman MK, Langenberg P, Bachorik PS et al. (2001). Non-high-density lipoprotein cholesterol level as a predictor of cardiovascular disease mortality. Arch Intern Med 61, 1413–1419.
Denke MA, Sempos CT, Grundy SM (1993). Excess body weight. An underrecognized contributor to high blood cholesterol levels in white American men. Arch Intern Med 153, 1093–1103.
Denke MA, Sempos CT, Grundy SM (1994). Excess body weight. An underrecognized contributor to dyslipidemia in white American women. Arch Intern Med 154, 401–410.
Despres JP (1994). Dyslipidemia and obesity. Balliere's Clin Endocrinol Metab 8, 629–660.
Despres JP, Lemieux I, Prud’homme D (2001). Treatment of obesity: need to focus on the high risk abdominally obese patients. BMJ 322, 716–720.
Farwell WR, Sesso HD, Buring JE, Gaziano JM (2005). Non-high-density lipoprotein cholesterol as a risk factor for a first non-fatal myocardial infarction. Am J Cardiol 96, 1129–1134.
Forsen T, Eriksson J, Qiao Q, Tervahauta M, Nissinen A, Tuomilehto J (2000). Short stature and coronary heart disease: a 35-year follow-up of the Finnish cohorts of The Seven Countries Study. J Intern Med 248, 326–332.
Frayn KN (2002). Adipose tissue as abuffer for daily lipid flux. Diabetologia 45, 1201–1210.
Gunnell D, Whitley E, Upton MN, McConnachie A, Davey Smith Gand G, Watt CM (2003). Associations of height, leg length, and lung function with cardiovascular risk factors in the Midspan Family Study. J Epidemiol Comm Health 57, 141–146.
Hebert PR, Rich-Edwards JW, Manson JE, Ridker PM, Cook NR, O’Connor GT et al. (1993). Height and incidence of cardiovascular disease in male physicians. Circulation 88, 1437–1443.
Henriksson KM, Lindblad U, Agren B, Nilsson-Ehle P, Rastam L (2001). Associations between body height, body composition and cholesterol levels in middle-aged men. The Coronary Risk Factor study in Southern Sweden (CRISS). Eur J Epidemiol 17, 521–526.
Hu D, Hannah J, Gray RS, Jablonski KA, Henderson JA, Robbins DC et al. (2000a). Effects of obesity and body fat distribution on lipids and lipoproteins in nondiabetic American Indians: the Strong Heart Study. Obes Res 8, 411–421.
Hu FB, Wang B, Chen C, Jin Y, Yang J, Stampfer MJ et al. (2000b). Body mass index and cardiovascular risk factors in a rural Chinese population. Am J Epidemiol 151, 88–97.
Hunter GR, Kekes-Szabo T, Snyder SW, Nicholson C, Nyikos I, Berland L (1997). Fat distribution, physical activity, and cardiovascular risk factors. Med Sci Sports Exerc 29, 362–369.
Innis SM (1985). The role of diet during development on the regulation of adult cholesterol homeostasis. Can J Physiol Pharmacol 63, 557–564.
Kanaya AM, Vittinghoff E, Shlipak MG, Resnick HE, Visser M, Grady D et al. (2003). Association of total and central obesity with mortality in postmenopausal women with coronary heart disease. Am J Epidemiol 158, 1161–1170.
Kannam J, Levy D, Larson M, Wilson P (1994). Short stature and risk for mortality and cardiovascular disease events. The Framingham Heart Study. Circulation 90, 2241–2247.
Khosla T, Lowe CR (1967). Indices of obesity derived from body weight and height. Br J Prev Soc Med 21, 122–128.
Kolacek S, Kapetanovic T, Zimolo A, Luzar V (1993). Early determinants of cardiovascular risk factors in adults. A. Plasma lipids. Acta Paediatr 82, 699–704.
Krahn AD, Manfreda J, Tate RB, Mathewson FA, Cuddy TE (1994). Evidence that height is an independent risk factor for coronary artery disease (the Manitoba Follow-Up Study). Am J Cardiol 74, 398–399.
La Batide-Alanore A, Tregouet DA, Sass C, Siest G, Visvikis S, Tiret L (2003). Family study of the relationship between height and cardiovascular risk factors in the STANISLAS cohort. Int J Epidemiol 32, 607–614.
Lawlor DA, Taylor M, Davey Smith G, Gunnell D, Ebrahim E (2004). Associations of components of adult height with coronary heart disease in postmenopausal women: the British women's heart and health study. Heart 90, 745–749.
Lissner L, Bjorkelund C, Heitmann BL, Seidell JC, Bengtsson C (2001). Larger hip circumference independently predicts health and longevity in a Swedish female cohort. Obes Res 10, 644–646.
Marmot MG, Page CM, Atkins E, Douglas JW (1980). Effect of breast-feeding on plasma cholesterol and weight in young adults. J Epidemiol Community Health 34, 164–167.
National Cholesterol Education Program Expert Panel. Third report of the National Cholesterol Education Program (NCEP). Expert Panel on Detection, Evaluation and Treatment of High Blood Cholesterol in Adults (ATP III) (2001). NIH Publication. National Heart, Lung and Blood Institute: Bethesda, MD.
National Task Force on the Prevention and Treatment of Obesity (2000). Overweight, Obesity, and Health Risk. Arch Intern Med 160, 898–904.
Palmer JR, Rosenberg L, Shapiro S (1990). Stature and the risk of myocardial infarction in women. Am J Epidemiol 132, 27–32.
Parker DR, Lapane KL, Lasater TM, Carleton RA (1998). Short stature and cardiovascular disease among men and women from two southeastern New England communities. Int J Epidemiol 27, 970–975.
Peck AM, Vagero DH (1989). Adult body height, self perceived health and mortality in the Swedish population. J Epidemiol Comm Health 43, 380–384.
Pi-Sunyer FX (2002). The Obesity epidemic: Pathophysiology and consequences of Obesity. Obes Res 10, 97S–104S.
Ravussin E, Smith SR (2002). Increased fat intake, impaired fat oxidation, and failure of fat cell proliferation result in fat storage, insulin resistance, and type 2 diabetes mellitus. Ann N Y Acad Sci 967, 363–378.
Riboli E, Hunt KJ, Slimani N, Ferrari P, Norat T, Fahey M et al. (2002). European Prospective Investigation into Cancer and Nutrition (EPIC) study populations and data collection. Public Health Nutr 5, 1113–1124.
Riboli E, Kaaks R (1997). The EPIC Project: Rationale and Study Design. Intern J Epidemiol 26 (Suppl 1), S6–S14.
Ridker PM, Rifai N, Cook NR, Bradwin G, Buring JE (2005). Non-HDL cholesterol, apolipoproteins A-I and B100, standard lipid measures, lipid ratios, and CRP as risk factors for cardiovascular disease in women. JAMA 294 (3), 326–333.
Rossman I (1977). Anatomic and body composition changes with age. In: Finch CE, Hayflick L (Eds). Handbook of the biology of aging. Van Nostrand Reinhold: New York, pp. 189–221.
Sakai Y, Ito H, Egami Y, Ohoto N, Hijii C, Yanagawa M et al. (2005). Favourable association of leg fat with cardiovascular risk factors. J Intern Med 257, 194–200.
Seidell JC, Pérusse L, Després JP, Bouchard C (2001). Waist and hip circumferences have independent and opposite effects on cardiovascular disease risk factors: the Quebec Family Study. Am J Clin Nutr 74, 315–321.
Shai I, Rimm EB, Hankinson SE, Curhan G, Manson JE, Rifai N et al. (2004). Multivariate assessment of lipid parameters as predictors of coronary heart disease among postmenopausal women: potential implications for clinical guidelines. Circulation 110, 2824–2830.
Sheehan MT, Jensen MD (2000). Metabolic complications of obesity. Pathophysiologic considerations. Med Clin North Am 84, 363–385.
Snijder MB, Zimmet PZ, Visser M, Dekker JM, Seidell JC, Shaw JE (2004a). Independent association of hip circumference with metabolic profile in different ethnic groups. Obes Res 12, 1370–1374.
Snijder MB, Zimmet PZ, Visser M, Dekker JM, Seidell JC, Shaw JE (2004b). Independent and opposite associations of waist and hip circumferences with diabetes, hypertension and dyslipidemia: the AusDiab Study. Int J Obes Relat Metab Disord 28, 402–409.
Svanborg ES, Mellstrom D (1991). Metabolic changes in aging: predictors of diseases. The Swedish experience. In: Ingram DK, Baker GT, Shock NW (Eds). The potential of nutritional modulation of aging. Food and Nutrition Press: Trumbull, CT, pp. 81–90.
Takada D, Ezura Y, Ono S, Iino Y, Katayama Y, Xin Y et al. (2003). Growth hormone receptor variant (L526I) modifies plasma HDL cholesterol phenotype in familial hypercholesterolemia: intra-familial association study in an eight-generation hyperlipidemic kindred. Am J Med Genet A 121, 136–140.
Thomas GN, Ho SY, Lam KS, Janus ED, Hedley AJ, Lam TH, Hong Kong Cardiovascular Risk Factor Prevalence Study Steering Committee (2004). Impact of obesity and body fat distribution on cardiovascular risk factors in Hong Kong Chinese. Obes Res 12, 1805–1813.
Wahrenberg H, Hertel K, Leijonhufvud BM, Persson LG, Toft E, Arner P (2005). Use of waist circumference to predict insulin resistance: retrospective study. BMJ 330, 1363–1364.
Wamala SP, Mittleman MA, Horsten M, Schenck-Gustafsson K, Orth-Gomer K (1999). Short stature and prognosis of coronary heart disease in women. J Intern Med 245, 557–563.
Willett WC, Stampfer MJ (1986). Total energy intake: implications for epidemiologic analyses. Am J Epidemiol 124, 17–24.
Williams MJ, Hunter GR, Kekes-Szabo T, Snyder S, Treuth M (1997). Regional fat distribution in women and risk of cardiovascular disease. Am J Clin Nutr 65, 855–860.
Yarnell JW, Limb ES, Layzell JM, Baker IA (1992). Height: a risk marker for ischaemic heart disease: prospective results from the Caerphilly and Speedwell Heart Disease Studies. Eur Heart J 13, 1602–1605.
Guarantor: A Trichopoulou.
Contributors: VB has initiated and implemented this study. CB was the lead biostatistician. DT is consultant epidemiologist. AT is the principal investigator in the Greek EPIC project. All authors have contributed to the drafting of the manuscript and have approved the final version.
About this article
Cite this article
Benetou, V., Bamia, C., Trichopoulos, D. et al. Associations of anthropometric characteristics with blood cholesterol fractions among adults. The Greek EPIC study. Eur J Clin Nutr 60, 942–948 (2006). https://doi.org/10.1038/sj.ejcn.1602403
- height, body mass index
- waist circumference
- hip circumference
Relationship Between Modified Body Adiposity Index and A Body Shape Index with Biochemical Parameters in Bariatric Surgery Candidates
Obesity Surgery (2020)
International Journal of Endocrinology (2016)
Identification of the Best Anthropometric Predictors of Serum High- and Low-Density Lipoproteins Using Machine Learning
IEEE Journal of Biomedical and Health Informatics (2015)
Association of height and pubertal timing with lipoprotein subclass profile: Exploring the role of genetic and environmental effects
American Journal of Human Biology (2013)
Non–High-Density Lipoprotein Cholesterol and Its Correlation With Anthropometric Markers of Cardiovascular Risk in Hemodialysis
Journal of Renal Nutrition (2012)