The overall participation rate of hypertensives was 86.5% (261 men, 258 women) and that of controls 87.7% (259 men, 267 women). The women were about one year older, because they were examined about one year later than the men. Twenty-six men and 32 women among the control subjects were on chronic medication affecting blood pressure, since BP-lowering medication is also commonly used for other indications apart from hypertension (e.g. chest pain, ankle edema, essential tremor, palpitations, etc.). Further, in a smaller portion (7 men and 14 women) of these 58 subjects, these drugs were used to treat hypertension, although the subjects were not listed as special refund recipients at the time of randomization, because they had been entitled to the refund after the randomization or had failed to apply for the right to refund for some other reason. The percentages of hypertensive subjects on medication for cardiovascular diseases were 93.9% of hypertensive men and 100% of hypertensive women. The prevalence rate was 58.6% for hypertensive men using beta-blockers, 37.2% for diuretic users, 23.8% for calcium channel blockers and 46.4% for angiotensin-converting enzyme inhibitors. For hypertensive women, the corresponding rates were 57%, 48.1%, 25.2% and 38%.
Table 2 shows the basic characteristics and cardiovascular risk factors in the control and hypertensive cohorts. Both hypertensive men and women were significantly more obese and had higher WH ratios than their age- and sex-matched controls. The systolic and diastolic blood pressures showed poor blood pressure control, especially in hypertensive men, of whom 70.1% exceeded the value of 160/95 mmHg (WHO definition for hypertension, Table 3). Both control and hypertensive men also had significantly higher blood pressures than the respective female cohorts.
Table 2. characteristics and cardiovascular risk factors of the hypertensive and control cohorts
| Variable | Control men | Hypertensive men | Control women | Hypertensive women |
|---|---|---|---|---|
| Number of patients | 259 | 261 | 267 | 258 |
| Age (yrs) | 50.9 (6.1) | 50.5 (5.9) | 51.8 (6.0) | 51.8 (5.9) |
| Height (cm) | 175.1 (6.6) | 174.9 (6.2) | 161.0 (5.7) | 160.8 (6.0) |
| Weight (kg) | 81.5 (12.9) | 90.0 (14.8)* | 67.9 (11.7) | 74.3 (14.7)* |
| BMI (kg/m2) | 26.5 (3.5) | 29.4 (4.4)† | 26.2 (4.4) | 28.7 (5.3)† |
| WH ratio | 0.91 (0.06) | 0.95 (0.06)* | 0.79 (0.06) | 0.82 (0.06)* |
| Systolic BP (mmHg) | 147.1 (19.8)‡ | 160.2 (20.3)‡ | 139.3 (21.1)‡ | 153.8 (20.5)‡ |
| Diastolic BP (mmHg) | 89.0 (10.7)‡ | 98.0 (10.0)‡ | 82.4 (12.2)‡ | 91.2 (11.1)‡ |
| Heart rate (beats/min) | 72 (15) | 73 (14) | 75 (11) | 76 (13) |
| Total cholesterol (mmol/L) | 5.78 (1.09)§ | 5.77 (1.02 | 5.53 (1.03) | 5.73 (1.07) |
| LDL cholesterol (mmol/L) | 3.73 (0.96)§ | 3.60 (0.92) | 3.32 (0.92) | 3.47 (0.92) |
| HDL cholesterol (mmol/L) | 1.22 (0.30) | 1.18 (0.31) | 1.56 (0.38)‡ | 1.43 (0.38)‡ |
| Triglycerides (mmol/L) | 1.57 (0.81)§ | 1.98 (1.26)‡ | 1.19 (0.68) | 1.60 (1.09)* |
| VLDL cholesterol (mmol/L) | 0.46 (0.33) | 0.59 (0.46)‡ | 0.31 (0.26)‡ | 0.44 (0.41)* |
| VLDL triglycerides (mmol/L) | 0.87 (0.59) | 1.17 (0.90)‡ | 0.56 (0.45) | 0.84 (0.71) |
| Alcohol consumption (absolute g/week) | 90 (100)§ | 106 (126)§ | 23 (36) | 31 (48) |
| IGT/DM (%) | 10.0/4.6 | 21.1†/16.5† | 13.9/3.8 | 19.8/12.8† |
| Smoking (pack-years) | 24.6 (11.0) | 22.6 (11.0) | 21.8 (11.1) | 22.0 (10.9) |
| Smokers (%) | 39.4‡ | 28.4 | 27.3 | 22.5 |
| fB-glucose (mmol/L) | 4.58 (1.05) | 5.14 (1.90)† | 4.34 (0.57) | 4.93 (1.82)† |
| 2-hr glucose (mmol/L) | 5.30 (2.34) | 6.62 (3.41)† | 5.43 (2.04) | 6.69 (3.61)† |
| fS-insulin (mU/L) | 13.2 (12.2) | 17.9 (12.9)‡ | 9.7 (6.2)‡ | 13.8 (10.3) |
| 2-hr insulin (mU/L) | 54.3 (49.5) | 86.8 (85.5)† | 57.8 (49.5) | 86.5 (95.0)† |
| ISI (mg·L2/(mmol·mU·min) | 101 (81, 123) | 77 (56, 100)† | 103 (81, 128) | 86 (59, 108)† |
| IRI (IR unit) | 2.78 (2.97) | 4.38 (4.47)† | 1.92 (1.47) | 3.24 (3.35)* |
| AUCINS (mU/L·h) | 98 (62, 163) | 138 (87, 215)† | 89 (63, 136) | 117 (75, 187)† |
| γ -GTP (U/L) | 46.5 (57.5)§ | 69.8 (110.2)‡ | 27.1 (18.8) | 45.8 (59.6)§* |
| Uric acid (µmol/L) | 285 (79)‡ | 338 (91)‡ | 201 (64)‡ | 250 (83)‡ |
| ALAT (U/L) | 32 (19) | 42 (33)‡ | 24 (13)* | 30 (18)§ |
| The values are mean (SD) except for ISI (Cederholm & Wibell 1990) and AUCINS, which are medians (25, 75 interquartile range); LDL, low-density lipoprotein; HDL, high-density lipoprotein; BP, blood pressure; IGT, impaired glucose tolerance; DM, diabetes mellitus; All statistical differences are P < 0.05; *cohort vs control cohort; †cohort vs controls in both sexes; ‡cohort vs all other cohorts; §cohort vs respective cohort of other sex. | ||||
The crude prevalence rates of the characteristics of the metabolic syndrome are shown in Table 3. By all the definitions of elevated blood pressure, control women had the lowest prevalence rates and hypertensive men the highest prevalence rates in spite of anti-hypertensive medication. Obesity (defined BMI > 25 kg/m2 or BMI > 30 kg/m2) was seen in 69.4% and 26.6% , respectively, of all subjects. A high WH ratio was present in 27.7% of all subjects. Both a high BMI and a high WH ratio were at least twice as common among hypertensive men than control women. The same phenomenon was visible in the fasting insulin levels. The prevalence rates of high cholesterol levels and low HDL cholesterol levels did not differ significantly in the different cohorts. Hypertensive men had markedly more often impaired glucose tolerance and diabetes mellitus than the control cohorts, either men or women. High triglyceride concentration was most prevalent in hypertensive men, among whom hypertriglyceridemia was almost 5 times more frequent than among control women.
Table 3. Prevalence of the characteristics of the metabolic syndrome in the hypertensive and control cohorts
| Variable | Control men (N = 259) | Hypertensive men (N = 261) | Control women (N = 267) | Hypertensive women (N = 258) |
|---|---|---|---|---|
| BMI > 25 (kg/m2) | 64.5 | 85.1† | 51.7§ | 76.7‡ |
| BMI > 30 (kg/m2) | 16.6 | 41.0† | 15.4 | 33.7† |
| WH ratio > 0.95M/0.85F | 22.8 | 45.2* | 12.4 | 30.2‡ |
| Diastolic BP > 95 mmHg | 30.1 | 65.1‡ | 15.0§ | 36.1‡ |
| Diastolic BP > 100 mmHg | 14.7 | 38.7‡ | 7.9§ | 18.6‡ |
| Diastolic BP > 110 mmHg | 2.7 | 11.1‡ | 0.8 | 5.4‡ |
| BP ≥ 160/95 mmHg | 37.8 | 70.1‡ | 21.4§ | 45.7‡ |
| fS-insulin (≥ 12 mU/L) | 39.4 | 65.1‡ | 24.7§ | 45.4‡ |
| Total cholesterol < 5.0 (mmol/L) | 23.9 | 24.1 | 30.7 | 24.4 |
| Total cholesterol 5.0-6.49 (mmol/L) | 53.7 | 54.4 | 53.2 | 55.0 |
| Total cholesterol 6.5 (mmol/L) | 22.4 | 21.5 | 16.1 | 20.5 |
| HDL cholesterol < 0.9M/1.1F (mmol/L) | 13.9 | 13.4 | 11.2 | 17.4 |
| Impaired glucose tolerance | 10.0 | 21.1† | 13.9 | 19.8 |
| Diabetes mellitus | 4.6 | 16.5† | 3.8 | 12.8† |
| Triglycerides ≥ 2.0 (mmol/L) | 20.1 | 35.3‡ | 7.1§ | 21.7‡ |
| The values are the actual, crude prevalence rates of the cohort (in percent); BP, blood pressure; HDL, high-density lipoprotein; WH, waist-to-hip ratio; F, females; M, males; All statistical differences are P < 0.05, (χ2-test); *cohort vs control cohort; †cohort vs controls in both sexes; ‡cohort vs control and respective cohort of other sex; §cohort vs control men. | ||||
The crude prevalence rates of all the abnormalities in their isolated form, i.e. associated with none of the other eight, were less than tenth of the overall prevalence rates. Two-by-two associations of isolated conditions were even less prevalent (data not shown, the prevalence rates ranging from 0% to 3.0%. Only the prevalence of the hypertension and obesity association as an isolated form reached up to 7.1%) and in each case less frequent than was to be expected by chance. As a consequence, almost all of the metabolic abnormalities occurred in combination with the others. However, control women had more often pure isolated conditions than the hypertensive cohorts, especially obesity, hyperinsulinaemia and hypercholesterolemia. If the lower limit for obesity was set at BMI > 30 kg/m2, only 17.5% of all the study subjects were free of all the nine disorders. The most marked difference was seen between the hypertensive men and the control women, the latter of whom were 8 times more often free of all the nine disorders (p < 0.0001).
The crude prevalence rates of the metabolic syndrome by different definitions in the hypertensive and control cohorts are shown in Table 4. The prevalence rates of the metabolic syndrome by the different definitions ranged from 0.8% to 35.3%. The most prevalent combination was hypertension and hypertriglyceridemia, which was seen in up to 35.3% of hypertensive men versus 2.6% of control women (p < 0.0001). The WHO criteria for the metabolic syndrome (Alberti & Zimmet 1998) without measures of obesity yielded prevalence rates of 5.2% for control women, 14.7% for control men, 26.4% for hypertensive women and 28.7% for hypertensive men. The characteristic feature was the tendency towards high prevalence rates among hypertensive men compared to the lowest prevalence rates among control women. Smoking did not have a significant relationship with the prevalence of the metabolic syndrome, although hypertensive women showed a nearly significant trend towards a higher prevalence rate of the metabolic syndrome in smokers (14% vs 25.9%, p = 0.052, respectively). The prevalence of the metabolic syndrome did not vary with menopause in control and hypertensive women.
Table 4. Prevalence of the metabolic syndrome by different definitions in the hypertensive and control cohorts
| Variable | Control men (N = 259) | Hypertensive men (N = 261) | Control women (N = 267) | Hypertensive women (N = 258) |
|---|---|---|---|---|
| BP + TRIG | 10.0 | 35.3‡ | 2.6§ | 21.7‡ |
| BP + INS +TRIG | 6.9 | 26.8‡ | 1.9§ | 16.7‡ |
| BP + INS + TRIG + HDL | 1.9 | 8.1† | 0.8 | 8.5† |
| BP + IGT/DM + TRIG | 2.3 | 16.9† | 1.9 | 12.4† |
| INS + TRIG + HDL | 4.6 | 8.1 | 2.3 | 8.5* |
| INS + HDL | 8.1 | 10.3 | 5.6 | 12.4* |
| INS + TRIG | 13.5 | 26.8‡ | 4.5§ | 16.7‡ |
| TRIG + HDL | 6.2 | 10.0 | 3.4 | 9.7* |
| TRIG + IGT/DM + INS | 3.9 | 12.3† | 3.4 | 9.3† |
| TRIG + IGT/DM + INS (mean + SD) | 0.8 | 3.8‡ | 2.6 | 8.5‡ |
| BP+IGT/DM+INS+HDL+TRIG (WHO) | 14.7 | 28.7† | 5.2§ | 26.4† |
| The values are the actual, crude prevalence rates of the cohort (in percent); BP, blood pressure ≥ 160/95 mmHg; No BP limits for hypertensives; INS, fS-insulin ≥ 12 mU/L; TRIG, triglycerides ≥ 2.0 mmol/L; HDL, high-density lipoprotein, men < 0.9 mmol/L, women < 1.1 mmol/L; IGT/DM, impaired glucose tolerance + diabetes mellitus; All statistical differences are P < 0.05, (χ2-test); *cohort vs control cohort; †cohort vs controls in both sexes; ‡cohort vs control and respective cohort of other sex; §cohort vs control men. WHO (Alberti & Zimmet 1998) | ||||
Multiple logistic regression analyses were performed to determine the independent predictors of the metabolic syndrome after adjustment for age, obesity and gender (Table 5). Because HDL cholesterol correlated inversely with triglycerides in the regression analyses, inducing a colinearity effect (odds ratio 0.03), it was excluded from the final analysis. Furthermore, low HDL cholesterol levels are included in the concept of MS. Of all the measures of obesity used in the analyses (BMI, waist circumference and WH ratio), waist circumference was the most powerful predictor of the metabolic syndrome in the different groups.
Table 5. Multiple logistic models for independent variables of the metabolic syndrome adjusted for age, measure of obesity in each model and sex in all study subjects (n= 1045)
| Model 1 | Model 2 | Model 3 | |||||
|---|---|---|---|---|---|---|---|
| Independent Variable | Odds Ratio (with a 95% confidence interval) | Independent Variable | Odds Ratio (with a 95% Confidence interval) | Independent variable | Odds Ratio (with a 95% confidence interval) | ||
| Waist (cm) | 1.05 (1.01–1.09) | BMI (kg/m2 ) | 1.18 (1.13–1.23) | WH ratio | 1.08 (1.03–1.12) | ||
| Uric acid (µmol/L) | 1.004 (1.002–1.007) | Uric acid (µmol/L) | 1.006 (1.003–1.008) | Uric acid (µmol/L) | 1.006 (1.003–1.008) | ||
| Total cholesterol (mmol/L) | 1.36 (1.14–1.63) | Total cholesterol (mmol/L) | 1.44 (1.19–1.72) | Total cholesterol (mmol/L) | 1.39 (1.17–1.66) | ||
| γ -GTP (U/L)* | 6.96 (3.81–12.75) | γ -GTP (U/L)* | 4.96 (2.64–9.33) | S-ALAT (U/L)* | 1.01 (1.005–1.02) | ||
| Metabolic syndrome defined as blood pressure ≥ 160/95 mmHg and fS-insulin ≥ 12 mU/L and fasting plasma triglycerides ≥ 2.0 mmol/L*γ -GTP was log10 transformed before analysis due to skewed distribution. | |||||||
To evaluate the attributable risk caused by the metabolic syndrome, we used the probability function based on the Framingham data (Wilson et al. 1987). The average calculated relative risk for coronary heart disease over the next six years was 1.6 for males (36% risk of CHD in six years) and 3.0 for females (3% risk of CHD in six years) with the metabolic syndrome. The risk of CHD in six years for a healthy male aged 50 with a total cholesterol of 5 mmol/L, HDL cholesterol of 1.22 mmol/L (mean value for control men), systolic blood pressure of 140 mm Hg and no other abnormalities is 3.9%, and that for a corresponding female 0.4% (HDL cholesterol 1.56 mmol/L, mean for control women).