Variation at the cholesteryl ester transfer protein gene in relation to plasma high density lipoproteins cholesterol levels and carotid intima-media thickness.

BACKGROUND: Cholesteryl ester transfer protein (CETP) plays a major role in lipoprotein metabolism. We have screened the CETP gene for mutations and polymorphisms regulating high density lipoproteins cholesterol (HDL-C) levels and the development of atherosclerosis, and found some polymorphisms (I405V and R451Q) to have minor effects. DESIGN: The purpose of this study was to investigate the combined effect of the several polymorphisms of the CETP gene so far found on HDL-C levels and carotid intima-media thickness (IMT), and, in addition, to study whether the recently found functional polymorphism in the promoter region of the CETP gene (C to A, - 629 relative to the first transcribed nucleotide) explains the previous associations due to linkage disequilibrium. The genotypes were determined in a population sample of 481 men and women. RESULTS: There were no significant differences in plasma CETP activity or carotid IMT between the genotypes of the promoter polymorphism. The women with the CC genotype of the promoter polymorphism had the lowest HDL-C levels (P < 0.001), but no such difference was seen in men. Detected polymorphisms of the CETP gene explained about 8% of the variation in HDL-C in women and about 7 and 10% of the variation in carotid IMT in women and men, respectively. The associations of the promoter, I405V and R451Q-A373P polymorphisms with HDL-C and carotid IMT seemed to be independent of each other. The associations with IMT were independent of total HDL-C levels, suggesting that HDL subfractions may have more effect on IMT. CONCLUSION: The CETP gene locus was found to be polymorphic and its polymorphisms explained a reasonable proportion of the variation in the degree of carotid atherosclerosis.

Effect of hypertension and risk factors on diameters of abdominal aorta and common iliac and femoral arteries in middle-aged hypertensive and control subjects: a cross-sectional systematic study with duplex ultrasound.

There is a general tendency towards atherosclerosis and arterial dilatation in older age, and high blood pressure also tends to increase arterial diameters. The purpose of this study was to examine the effect of hypertension and other cardiovascular risk factors on aortic, common iliac and common femoral artery diameters. The diameters of the abdominal aorta and the iliac and femoral arteries and the extent of echogenic plaques in the aorta and the iliac arteries down to groin level were evaluated with ultrasound in 1007 middle-aged (40-60 years) men (505) and women (502), 496 with arterial hypertension and 511 controls. Twenty-eight subjects were excluded because of poor visualization. Men had significantly larger diameters of the abdominal aorta (mean 21.3+/-2.8 vs. 17.8+/-1.3 mm) and the common iliac (13.4+/-2.0 vs. 12.2+/-1.2) and common femoral arteries (11.0+/-1.4 vs. 9.7+/-0.9) than women (P for all <0.001), but arterial diameter was also related to the subject's size. Atherosclerotic plaques, age and height were associated with the diameter of the abdominal aorta in men, while high body mass index (BMI) had less significance. The diameter of the aorta was larger in hypertensive men aged 56-60 than in controls of the same age. In women, height, BMI and diastolic blood pressure (DBP) were associated with the diameter of the aorta, while systolic blood pressure (SBP) had less and age no effect. Age, plaques, height, BMI, DBP and SBP were associated with the diameters of the common iliac arteries in both genders, while smoking had an inverse correlation. The results on lipid values were inconsistent and an abnormal glucose tolerance test proved nonsignificant. In conclusion, arterial size measured as a diameter related to the subject's size was larger in men. Age, arterial plaques and blood pressure increased arterial diameter significantly. However, the hypertensive disease itself had only a minimal effect. The changes were smaller in women than in men.

Gamma-glutamyl transpeptidase and the metabolic syndrome.

OBJECTIVE: To analyse the associations between serum gamma-glutamyl transpeptidase activity (GTP) and the components of the metabolic syndrome. DESIGN: Cross-sectional, observational study of hypertensive patients and controls. SETTING: The participating subjects visited the research laboratory of the Department of Internal Medicine, University of Oulu, Oulu, Finland. SUBJECTS: A total of 1045 Caucasians, 40-59 years of age, consisting of 261 drug-treated hypertensive men, 258 drug-treated hypertensive women and 526 age- and sex-matched controls. MAIN OUTCOME MEASURES: The associations between GTP and the cardiovascular risk factors were analysed through multiple regression and logistic methods and by GTP tertiles. The independent effect of GTP on different insulin measures, calculated from the values of 2 h of oral glucose tolerance test, was estimated after concurrent adjustment for age, obesity and alcohol consumption. RESULTS: GTP correlated highly significantly with the components of the metabolic syndrome. The correlation coefficient were 0.33 between GTP and body mass index, 0. 25 between GTP and systolic blood pressure in control men (P = 0. 0001), 0.39 between GTP and triglycerides, and 0.32 between GTP and fasting insulin in hypertensive women (P = 0.0001). The association between GTP and blood pressure remained significant only at upright measurements in controls. All insulin measures had a significant positive association with increasing GTP tertiles in all the study groups (e.g. fasting insulin 8.1 mU L-1 in the lowest and 11.0 mU L-1in the highest tertile in control women, P = 0.0001), with the exception of fasting insulin in control men. In a pooled logistic analysis after adjustment for age, body mass index, alcohol consumption and gender, the independent predictors of the metabolic syndrome were body mass index, uric acid, total cholesterol and GTP (for log-transformed GTP odds ratio 4.0, 95% CI: 2.80-5.69). CONCLUSIONS: There are significant associations between GTP and the components of the metabolic syndrome. Elevated levels of GTP may not always indicate increased alcohol consumption, but may also suggest the existence of the metabolic syndrome with its subsequent deleterious consequences.

Dietary and other non-pharmacological treatments in patients with drug-treated hypertension and control subjects.

OBJECTIVE: To study the nutrient intakes and other lifestyle patterns of drug-treated hypertensives and control subjects. DESIGN: A cross-sectional population-based epidemiological study. SETTING: The participating study subjects visited the research laboratory of the Department of Internal Medicine of the University of Oulu, Oulu, Finland. PARTICIPANTS: A total of 1045 Finnish men and women aged 40-60 years, of whom 716 (69%) completed 7-day food records. MAIN OUTCOME MEASURES: Intakes of energy, protein, total fat, saturated, monounsaturated and polyunsaturated fatty acids, carbohydrate, alcohol, fibre, calcium, magnesium, potassium and sodium were assessed from 7-day food records. The study also included measurements of blood pressure, blood glucose and plasma lipids, and anthropometric variables. Information about alcohol consumption, smoking habits and physical activity was collected by interviewing. RESULTS: Obesity was common amongst the hypertensive subjects, whose body mass indices were significantly higher than those of the control subjects. Only minor differences existed in the energy intake levels and nutrient intakes of the hypertensive and control cohorts, but the control subjects performed more physical activity than the hypertensive subjects. The dietary recommendations concerning the intakes of total and saturated fats, carbohydrate and fibre were poorly met by both the hypertensive and the control subjects. Alcohol consumption was high amongst the hypertensive men, especially amongst the smokers. CONCLUSIONS: Non-pharmacological treatment, including dietary management, of hypertensive patients at high risk for cardiovascular complications seems still to be inadequate. Additional well-focused efforts are needed to intensify the dietary treatment as well as to reduce alcohol consumption and smoking amongst hypertensives.

Cholesteryl ester transfer protein gene polymorphisms are associated with carotid atherosclerosis in men.

BACKGROUND: The cholesteryl ester transfer protein (CETP) is involved in the reverse cholesterol transport and is therefore a candidate gene for atherosclerosis. DESIGN: The prevalences of the I405V and the R451Q polymorphisms were studied in a population sample of 515 men and women. Genotypes were determined by PCR and carotid atherosclerosis by ultrasonography as the mean intima-media thickness (IMT) of the carotid arteries. RESULTS: The Q451 allele was associated with significantly lower intima media thickness in men (P = 0.001). The Q451 allele was, in our earlier study, associated with high plasma CETP activity in men. The VV405 genotype was associated with lower plasma CETP activity compared with the II405 genotype (P < 0.01 for the difference). In the general linear model general factorial procedure the interaction between alcohol consumption and the I405V genotype on IMT was significant (P = 0.013) in men, and when the interaction term was taken into the model the I405V genotype also significantly affected IMT (P = 0.008). The VV405 genotype seems to be most harmful for men with the highest alcohol consumption. CONCLUSIONS: We describe two polymorphisms of the CETP gene associated with intima media thickness in men. A significant interaction was found between alcohol consumption and the I405V genotype on IMT.

Prevalence of the metabolic syndrome in drug-treated hypertensive patients and control subjects.

OBJECTIVES: To determine the prevalence of the metabolic abnormalities associated with hypertension and to define the predictors of the metabolic syndrome by different definitions in random population-based samples. DESIGN: A cross-sectional epidemiological study of hypertensive patients and controls. SETTING: The participating study subjects visited the research laboratory of the Department of Internal Medicine, University of Oulu, Oulu, Finland. SUBJECTS: Six hundred treated male and female hypertensives aged 40-59 years and 600 age- and sex-matched controls were randomly selected by age stratification from population registers. MAIN OUTCOME MEASURES: A wide range of laboratory analyses were conducted. After fasting blood had been drawn, the subjects were given a 75 g glucose load except previously known insulin-treated diabetics. Both 1 h and 2 h glucose and insulin concentrations were determined. During the same visit, a standardized health questionnaire covering the past medical history, current and former medication use, physical activity, smoking habits, alcohol consumption and family history was completed. Ten different definitions of the metabolic syndrome were applied to achieve a wide perspective of the prevalence of the different combinations. RESULTS: The prevalence of the metabolic syndrome in different samples varied depending on the definition from 0.8 to 35.3%, being lowest in the control men and women and highest in the hypertensive men. Three-quarters of a random, middle-aged, urban population show at least one cardiovascular risk factor and 91.3% of all the hypertensive subjects show at least one cardiovascular risk factor in addition to hypertension itself. The independent predictors of the metabolic syndrome were waist circumference, uric acid, total cholesterol and gamma-glutamyl transpeptidase in logistic analysis after adjustment for age, measure of obesity and gender. CONCLUSIONS: This cross-sectional, epidemiological study shows that the magnitude of the prevalence rates of the metabolic syndrome is at the same level in various populations, being less than one-third in population-based samples in spite of the different definitions. The cluster of several cardiovascular risk factors, especially in the hypertensives, leads to an increased relative risk of cardiovascular diseases.

Apolipoprotein E phenotype determines the effect of alcohol on blood pressure in middle-aged men.

Apolipoprotein E (apoE) has an essential role in lipoprotein metabolism, but recent studies have also revealed other functions associated with it, eg, neurologic and malignant diseases. We studied the association between apoE phenotypes E2/3, E3/3, and E4/3 and blood pressure after adjustment for covariates, as well as the association between phenotypes and adjusted plasma glucose and insulin levels in the standard oral glucose tolerance test in a random middle-aged population-based cohort of 259 men and 267 women. Systolic blood pressure was associated with apoE phenotype in the men with moderate or heavy alcohol consumption (>115 g/week), the mean systolic blood pressure value being 16 mm Hg higher in the E2/3 and 11 mm Hg higher in the E3/3 phenotypes than in the E4/3 phenotype, P = .04. No association was seen in occasional drinkers or teetotalers (lowest tertile <24 g/week), whereas in the middle tertile the association was intermediate. The same association was seen with diastolic blood pressure. In men, there was a significant correlation between systolic blood pressure and alcohol consumption in the E2/3 phenotype (rs = 0.71, P < .01) and in the E3/3 phenotype (rs = 0.25, P < .01), but not in the E4/3 phenotype (rs = 0.03, NS). No association between apoE phenotypes and insulin resistance was observed. In conclusion, in middle-aged men, apoE phenotype significantly influences the blood-pressure-increasing effect of alcohol consumption. This gene environment interaction may have marked implications for the prevention and treatment of hypertension.

Effect of hypertension, diabetes and other cardiovascular risk factors on kidney size in middle-aged adults.

The aim of the study was to investigate in a population-based series (1031 subjects, age range 40-60 years) whether the renal size of hypertensive subjects differs from that of control subjects and whether the difference might be due to hypertension itself or risk factors associated with hypertension. The renal measurements were performed by abdominal ultrasound. The genders were analyzed separately. Hypertensive men had slightly larger kidneys than controls (70.1+/-8.9 cm2 vs. 67.9+/-8.7 cm2, p <0.008). The difference was, however, mediated mainly through the body mass index (BMI), whereas hypertension, blood pressure or hypertensive medication did not affect renal size. High serum concentrations of uric acid and creatinine were associated with smaller kidney size (p <0.001 and p <0.05, respectively). Alcohol users had slightly larger kidneys than abstainers, but the difference was not significant. Renal size increased with pack years of smoking. Diabetics had 4.8% larger kidneys (p <0.039), but no difference was observed between the subjects with impaired glucose tolerance and those with normal test results. In multivariate analysis, the most significant factors associated with enlarged kidney size were the fasting blood glucose concentration (p < or = 0.0001), smoking (p < or = 0.0001) and atherosclerotic lesions in carotid arteries (p <0.002). The kidneys were also slightly larger in hypertensive women than in control subjects, but the difference was only of borderline significance (p <0.08). Women on hormone replacement therapy had smaller kidneys than other women (p <0.05), but there was no difference in renal measures between premenopausal and postmenopausal women. In multivariate analysis, the most significant factors contributing to large kidney size were blood glucose concentration (p <0.0001) and smoking (p <0.05), while age and serum creatinine concentration were associated with smaller kidney size (p <0.0001 and p <0.0001). We conclude that renal size is related to sex and the subject's height and weight. Smoking, abnormal glucose tolerance, blood uric acid, creatinine, carotid atherosclerosis and hormone replacement therapy in women were also significant factors for renal size. Hypertensive subjects had larger kidneys than controls, mainly because of their more frequent obesity and abnormal glucose test.

Variants of renin-angiotensin system genes and echocardiographic left ventricular mass.

AIMS: Variants of renin-angiotensin system genes are shown to be associated with cardiovascular pathology. The association between renin-angiotensin system genes and left ventricular mass was investigated in a population-based case-control study. METHODS AND RESULTS: The association between echocardiographic left ventricular mass and both insertion/deletion polymorphism of the angiotensin-converting enzyme gene and the methionine-threonine variant at position 235 of the angiotensinogen gene was studied in a random cohort of 430 hypertensive and 426 control subjects. No differences in the adjusted left ventricular mass values between the different genotypes were seen among either the hypertensive or the control subjects, whether men or women, or in the subgroups of normotensive or physically active subjects. Gene variation had no statistically significant synergistic effect on left ventricular mass values. In control women, the deletion allele of the angiotensin-converting enzyme gene was associated with an increased risk of left ventricular hypertrophy. However, this finding was based on a small number of women with left ventricular hypertrophy and should be interpreted with caution. CONCLUSION: Variations in renin-angiotensin system genes had no major effect on left ventricular mass in this middle-aged population-based cohort of hypertensives and control subjects.

Hyperinsulinemia and carotid atherosclerosis in hypertensive and control subjects.

OBJECTIVE: To analyze the relationships between carotid atherosclerosis measured as intima-media thickness (IMT) and different measures of insulin in a population-based case-control study of men and women. RESEARCH DESIGN AND METHODS: Carotid ultrasonographic measurements and 2-h oral glucose tolerance tests were performed in a random sample of 513 hypertensive subjects, aged 40-59 years, and in 518 age- and sex-matched control subjects. The associations between IMT and the different measures of insulin were analyzed through multiple regression and by insulin quintiles. The independent effect of insulin was estimated after concurrent adjustment for age, obesity, LDL cholesterol, and systolic blood pressure. RESULTS: The most powerful correlates with IMT were LDL cholesterol, age, systolic blood pressure, pack-years of smoking, and of the different insulin parameters, 2-h post-load insulin. In stepwise regression analysis, the independent predictors of the mean IMT were LDL cholesterol, systolic blood pressure, pack-years of smoking, and age (P < 0.0001) after adjustment for the independent predictors. In analysis of variance, no positive association of insulin parameters with IMT was found between the 2-h insulin quintiles after adjustment for the independent variables. The exclusion of diabetic subjects did not change the results. CONCLUSIONS: The present study of a population-based sample of men and women found inconsistent associations between different insulin measures and IMT after adjustment for the independent variables.

Heart rate variability and baroreflex sensitivity in hypertensive subjects with and without metabolic features of insulin resistance syndrome.

Both abnormal autonomic control of heart rate, assessed by heart rate variability (HRV) and baroreflex sensitivity (BRS), and insulin resistance syndrome are common in hypertensive patients. It is not known, however, whether abnormalities in HRV and BRS in hypertension are related to the insulin-resistance syndrome. Therefore, we compared HRV and BRS in hypertensive subjects with and without metabolic features of the insulin-resistance syndrome. HRV was analyzed using the autoregressive method from a 45-min electrocardiographic recording (15 min lying, sitting, and standing) and BRS using the Valsalva maneuver. The groups were matched for age, sex, and antihypertensive medication, and age- and sex-matched normotensive subjects served as a control group (n = 69 in each group). The insulin-resistance syndrome was defined using the criteria of 1) hypertension (blood pressure >160/90 mm Hg), 2) hypertriglyceridemia (fasting serum triglycerides > or =2.0 mmol/L), and 3) hyperinsulinemia (fasting serum insulin > or =12 mU/L). Standard deviation of RR intervals, total, very-low-, and low-frequency power of HRV were significantly lower in hypertensive subjects with insulin-resistance syndrome compared to hypertensive subjects without the syndrome and to normotensive controls (P < .001 for all), but the hypertensive group without the syndrome did not differ from the normotensive group. High-frequency power of HRV (P < .01) and BRS (P < .05) were reduced in both hypertensive groups compared to the normotensive group. In multiple regression analysis, systolic blood pressure (P < .01) and serum triglyceride level (P < .001) were independent predictors of reduced total power of HRV, but BRS was related only to systolic blood pressure (P < .01). Thus, most of the abnormalities in overall HRV seem to be confined to the subgroup of hypertensive subjects with insulin-resistance syndrome, but baroreflex and respiratory modulation of heart rate are impaired also in hypertensive subjects without metabolic features of insulin-resistance syndrome.

Variation at the angiotensin-converting enzyme gene and angiotensinogen gene loci in relation to blood pressure.

To investigate whether the polymorphisms in the angiotensin-converting enzyme and angiotensinogen genes are associated with hypertension, we carried out a case-control study of 508 hypertensive and 523 control subjects randomly selected from the Social Insurance Institution register. The cohorts were well characterized and matched for age and sex. The insertion/ deletion polymorphism of the angiotensin-converting enzyme gene and the methionine-->threonine variant at position 235 of the angiotensinogen gene were determined by the polymerase chain reaction technique. The allele frequencies and genotype distributions of both polymorphisms were similar in hypertensive and control subjects. Systolic and diastolic pressures adjusted for age, body mass index, and alcohol consumption did not differ significantly between the different genotypes of the angiotensin-converting enzyme and angiotensinogen genes. The variation at the angiotensinogen and angiotensin-converting enzyme genes did not have any statistically significant synergistic effect on blood pressure levels. In conclusion, the polymorphisms in the reninangiotensin cascade genes do not confer a significantly increased risk for the development of hypertension in this middle-aged, population-based cohort.

Association between angiotensin converting enzyme gene polymorphism and carotid atherosclerosis.

OBJECTIVE: Variations in the angiotensin converting enzyme (ACE) gene have been implicated in cardiovascular pathology. Therefore, the association between the intima-media thickness (IMT) of the carotid artery and the insertion/ deletion (I/D) polymorphism of the ACE gene was investigated. SUBJECTS: Three hundred men and 300 women were selected randomly from the middle-aged population living in the town Oulu, Finland, of whom 515 subjects (85.8%) participated. METHODS: The IMT of the carotid arteries was determined by bilateral B-mode ultrasonography. IMT values were adjusted for gender, age, height, plasma low-density lipoprotein cholesterol level, smoking and systolic blood pressure. The I/D polymorphism of the ACE gene was determined by polymerase chain reaction. RESULTS: Among non-smokers, the subjects with the DD genotype had significantly higher carotid IMT than did those with II or ID. The association was found also in combined IMT plaque values. In the total population the association was weaker and it was absent in current smokers. Genotype could explain 1.3-2.7% of the variance of carotid IMT in non-smokers. No association between the amount or size of carotid plaques and genotype was observed. CONCLUSIONS: Variations at the ACE gene locus contribute to the degree of the early changes in carotid atherosclerosis in the population. The gene effect is, however, masked by stronger effects of environmental factors such as smoking. The lack of association between atherosclerotic plaques and genotypes may reflect different mechanisms being involved in plaque development and early arterial wall thickening.

Sex-related differences in autonomic modulation of heart rate in middle-aged subjects.

BACKGROUND: Women have worse outcomes when they experience acute myocardial infarction (MI), but the reasons for this sex-related difference are not well understood. Because cardiovascular neural regulation plays an important role in cardiac mortality, we studied possible sex-related differences in the autonomic modulation of heart rate (HR) in middle-aged subjects without known heart disease. METHODS AND RESULTS: Baroreflex sensitivity (BRS) and HR variability were studied in randomly selected, age-matched populations of middle-aged women (n = 186; mean age, 50 +/- 6 years) and men (n = 188; mean age, 50 +/- 6 years) without hypertension, diabetes, or clinical or echocardiographic evidence of heart disease. BRS measured from the overshoot phase of the Valsalva maneuver was significantly lower in women (8.0 +/- 4.6 ms/mm Hg, n = 152) than in men (10.5 +/- 4.6 ms/mm Hg, n = 151) (P < .001), and the low-frequency component of HR variability measured from ECG recordings also was lower in women (P < .001), whereas the high-frequency component was higher in women than in men (P < .001). The ratio between the low-and high-frequency oscillations also was lower in the women (P < .001). The increase of HR and decrease of high-frequency component of HR variability in response to an upright posture were smaller in magnitude in women than in men (P < .01 for both). After adjustment for differences in the baseline-variables, such as blood pressure, HR, smoking, alcohol consumption, and psychosocial score, the sex-related differences in BRS and HR variability still remained significant (P < .001 for all). Women with estrogen replacement therapy (n = 46) had significantly higher BRS and total HR variance than the age-matched women without hormone treatment (P < .01 for both), and the BRS and HR variability of the women with estrogen therapy did not differ from those of the age-matched men. CONCLUSIONS: Baroreflex responsiveness is attenuated in middle-aged women compared with men, but the tonic vagal modulation of HR is augmented. Hormone replacement therapy appears to have favorable effects on the cardiovascular autonomic regulation in postmenopausal women.

Heart rate variability in systemic hypertension.

Low heart rate (HR) variability is a risk factor for cardiac mortality in various patient populations, but it has not been well established whether patients with long-standing hypertension have abnormalities in the autonomic modulation of HR. Time and frequency domain measures of HR variability were compared in randomly selected, age-matched populations of 188 normotensive and 168 hypertensive males (mean age 50 +/- 6 years for both). The standard deviation of the RR intervals was lower in the hypertensive subjects than in the normotensive ones (52 +/- 19 vs 59 +/- 20 mss; p <0.01), and the very low and low-frequency spectral components of HR variability analyzed as absolute units were reduced in the hypertensive patients relative to the normotensive controls (p <0.001 for both). Hypertensive subjects also had blunted changes of the normalized low- and high-frequency components in response to an upright (sitting) posture (NS) as compared with normotensive subjects (p <0.001 for both). Multiple regression analysis showed the standard deviation of the RR intervals to be predicted most strongly by systolic blood pressure, both in the patients with hypertension (beta--0.20, p=0.01) and in the normotensive subjects (beta--0.28, p=0.0002). After adjustment for the baseline differences in blood pressure and body mass index, none of the absolute measures of the HR variability or the responses of the normalized units of HR variability to a change in the body posture differed between the hypertensive subjects and normotensive controls. These data show that long-standing hypertension results in reduced overall HR variability and blunted autonomic responses to a change in body posture. Altered autonomic modulation of HR in hypertension is mainly due to elevated blood pressure and obesity in males with long-standing hypertension as compared with normotensive subjects.

Sex difference in the regulation of plasma high density lipoprotein cholesterol by genetic and environmental factors.

Association between high density lipoprotein (HDL) cholesterol concentration and restriction fragment length polymorphisms at the cholesteryl ester transfer protein (CETP) gene locus was studied in a random population-based cohort of 526 Caucasian subjects (259 men, mean age 50.9 years, and 267 women, mean age 51.8 years). HDL cholesterol concentration was adjusted for age, body mass index, alcohol consumption, smoking and plasma triglyceride and low density lipoprotein cholesterol levels. In females, the HDL cholesterol levels were associated with TaqIB polymorphism (1.46 mmol/l in the B1B1 genotype, 1.56 mmol/l in B1B2 and 1.72 mmol/l in B2B2, P = 0.0001 for the trend). In contrast, this was not observed in men (1.24, 1.20, 1.27 mmol/l, NS). The association was seen even in women who were current smokers (1.41, 1.56, 1.75 mmol/l, n = 72, P = 0.007), but not in male smokers (1.26, 1.19, 1.14 mmol/l, n = 102, NS). In male non-smokers the association was weak (1.22, 1.20, 1.32 mmol/l, n = 157, P = 0.05). In postmenopausal women not receiving hormone replacement therapy (n = 108), the association continued to be present, although weaker (1.50, 1.58, 1.70 mmol/l, P = 0.06). CETP activity (n = 101) tended to be lower in subjects with the B2B2 genotype. In conclusion, a clear-cut sex difference was observed in the genotype effect on plasma HDL cholesterol levels. The slight attenuation of the gene dosage effect after menopause suggests that the gender difference may be, at least in part, due to sex hormones. A genetic subgroup (men with the B2B2 genotype) particularly susceptible to the HDL cholesterol decreasing effect of smoking could be demonstrated.