The metabolic syndrome (MS) is characterized by 4 independent risk factors for cardiovascular disease: elevated triglyceride-rich lipoproteins, reduced high-density lipoproteins, elevated blood pressure, and dysglycemia. Several underlying risk factors, notably obesity, accentuate these independent risk factors. This study addressed 2 questions: Is the prevalence of MS identified equally by all measures of obesity? and Should any measure of obesity be included among diagnostic components of the MS? A cohort of 8,879 women and 23,145 men in the Cooper Center Longitudinal Study (CCLS) underwent anthropometric assessment and risk-factor measurement. Most subjects were Caucasian, and 13.1% of women and 30.5% of men had MS defined by the National Cholesterol Education Program Adult Treatment Panel (ATP) III guidelines. In ATP III, MS is diagnosed by any 3 of 5 factors (i.e., the 4 independent risk factors listed previously plus abdominal obesity, defined as increased waist girth). In the CCLS, several measures of obesity (e.g., percentage body fat, body mass index, and truncal subcutaneous fat) were found to substitute for elevated waist girth without appreciably changing MS prevalence. The impact of removing obesity from the diagnostic criteria (abridged ATP III MS, defined as 3 of 4 independent risk factors) was further examined. Abridged ATP III MS was less common than ATP III MS but recognized a subgroup of patients at higher risk for cardiovascular disease. In conclusion, abridged ATP III MS appears to be preferable to ATP III MS for the detection of candidates for intensive cardiovascular risk reduction.
The metabolic syndrome (MS) is a multiplex risk factor for cardiovascular disease (CVD) consisting of independent risk factors (elevations in triglyceride-rich lipoproteins, glucose, and blood pressure plus reduced high-density lipoprotein [HDL]), emerging risk factors (e.g., prothrombotic and proinflammatory states), and underlying risk factors (e.g., obesity, physical inactivity, and genetic and endocrine abnormalities). The latter contribute to independent and emerging risk factors. By consensus, a diagnosis of MS can be made from findings of any 3 of 5 factors: elevations of triglycerides, blood pressure, fasting glucose, and waist girth (WG) as well as low HDL cholesterol. These particular criteria for diagnosis were first proposed by the National Cholesterol Education Program Adult Treatment Panel (ATP) III. They have been modified slightly in a recent worldwide consensus definition. Recently, Franco et al compared the risk associated with various combinations of 3 risk factors in the Framingham Offspring Study, which raised the question of whether different treatment strategies should be used for different sets of risk factors associated with a diagnosis of MS. In the present study, we examined 2 additional questions: (1) whether the measurement of WG is superior to other measurements of obesity in the diagnosis of MS and (2) whether obesity should be counted at all among the diagnostic criteria. These questions were addressed in the large population of the Cooper Center Longitudinal Study (CCLS).
Methods
The CCLS (formerly called the Aerobics Center Longitudinal Study) was originally designed to examine relations between physical activity and cardiorespiratory fitness and health outcomes. The institutional review board at the Cooper Institute has approved the study annually since its inception in 1970. Subjects underwent clinical histories, physical examinations, anthropometry, cardiovascular fitness testing, and evaluations for cardiovascular risk factors. Data from the first clinic visit for subjects participating from 1979 to 2008 are included. Adult women (n = 8,879) and men (n = 23,145) who underwent anthropometry and assessments of fasting blood lipids, lipoprotein cholesterol, blood glucose, and blood pressure were selected for an ad hoc analysis of the association between body composition and MS risk factors. Anthropometric measures included height, weight, waist and hip girths, and skinfold thicknesses.
The baseline examination was conducted after a 12-hour fast. Body mass index (BMI) was computed from height and weight. WG was measured at the level of the umbilicus and hip girth at the widest extension of the buttocks after exhalation. Skinfold measurements at 7 sites were used to calculate percentage body fat. Details of sample analysis have been published previously.
In this study, standard BMI categories of lean (<25 kg/m 2 ), overweight (25.0 to 29.9 kg/m 2 ), and obese (≥30 kg/m 2 ) were synchronized to WG. To determine the WG equivalence to BMI for each gender, linear regression lines were derived between BMI and WG. The regression equations were WG (cm) = 2.1 × BMI + 24.4 (r = 0.88, p <0.0001) for women and WG (cm) = 2.4 × BMI + 28.6 (r = 0.89, p <0.0001) for men. The following WG ranges were found to be equivalent to BMI categories: lean, WG <77.5 cm (women) and <89 cm (men); overweight, WG 77.5 to 88.2 cm (women) and 89 to 101.1 cm (men); and obese, WG >88.2 cm (women) and >101.1 cm (men).
ATP III MS was defined by a slightly simplified version of ATP III criteria, that is, ≥3 of the following 5 criteria: fasting glucose ≥100 mg/dl, fasting triglycerides ≥150 mg/dl, HDL ≤40 mg/dl for men and ≤50 mg/dl for women, blood pressure ≥130/85 mm Hg, and WG >101.1 cm for men and >88.2 cm for women. Abridged ATP III MS was defined as the presence of ≥3 of 4 criteria, which excludes girth or any of its obesity equivalents.
Data are summarized as mean ± SD across WG categories. Trends across body composition categories were tested to evaluate differences in anthropometric measures, lipids, and systolic blood pressure with regression analysis. Spearman’s correlations coefficients were calculated for anthropometric measures compared to metabolic and systolic blood pressure parameters. Partial correlations were estimated after normalizing for percentage body fat. An α value <0.05 was considered significant in the data analyses. All analyses were conducted using SAS version 9.1 (SAS Institute Inc., Cary, North Carolina).
Results
The anthropometric characteristics of study subjects are listed in Table 1 . They were divided into lean, overweight, and obese categories defined by WG cut points as calculated using the WG regression equations given previously. Women fell mainly into the lean category. WGs and waist-to-hip ratios were greater on average in men than in women, but average percentage body fat was lower in men than women in each girth category.
| Variable | WG Category † | ||
|---|---|---|---|
| Lean | Overweight | Obese | |
| Women | |||
| Number of subjects | 5,401 | 2,073 | 1,405 |
| Age (years) | 45 ± 10 | 49 ± 11 ‡ | 50 ± 11 ‡ |
| BMI (kg/m 2 ) | 21.8 ± 2.3 | 26.4 ± 2.6 ‡ | 32.5 ± 5.4 ‡ , § |
| Body fat (%) ∥ | 22.4 ± 5.3 | 29.9 ± 4.1 ‡ | 33.9 ± 4.4 ‡ , § |
| Waist girth (cm) | 69.2 ± 4.7 | 82.3 ± 3.1 ‡ | 98.4 ± 8.8 ‡ , § |
| Waist-to-hip ratio | 0.74 ± 0.04 | 0.79 ± 0.06 ‡ | 0.88 ± 0.14 ‡ , § |
| Truncal fat ⁎ (mm) | 14.7 ± 5.1 | 23.4 ± 5.0 ‡ | 30.2 ± 7.2 ‡ , § |
| Thigh fat ⁎ (mm) | 28.3 ± 8.4 | 34.9 ± 9.9 ‡ | 41.9 ± 12.4 ‡ , § |
| Truncal fat/thigh fat ratio | 0.55 ± 0.21 | 0.72 ± 0.27 ‡ | 0.78 ± 0.4 ‡ , § |
| Men | |||
| Number of subjects | 6,381 | 10,572 | 6,097 |
| Age (years) | 43 ± 10 | 47 ± 10 ‡ | 49 ± 10 ‡ |
| BMI (kg/m 2 ) | 23.9 ± 1.9 | 27.1 ± 2.1 ‡ | 32.5 ± 4.2 ‡ , § |
| Body fat (%) ∥ | 16.7 ± 4.3 | 22.4 ± 3.6 ‡ | 28.4 ± 4.0 ‡ , § |
| Waist girth (cm) | 83 ± 4.2 | 94.7 ± 3.6 ‡ | 110.4 ± 8.7 ‡ , § |
| Waist-to-hip ratio | 0.86 ± 0.04 | 0.92 ± 0.04 ‡ | 0.98 ± 0.05 ‡ , § |
| Truncal fat ⁎ (mm) | 15.9 ± 4.8 | 22.8 ± 4.7 ‡ | 30.7 ± 6.5 ‡ , § |
| Thigh fat ⁎ (mm) | 14.2 ± 5.2 | 16.9 ± 6.0 ‡ | 22.2 ± 9.1 ‡ , § |
| Truncal fat/thigh fat ratio | 1.22 ± 0.47 | 1.48 ± 0.52 ‡ | 1.57 ± 0.59 ‡ , § |
⁎ Skinfold thickness measured with calipers at 7 sites (axilla, chest, abdomen, back, triceps, thigh, and hip). Average truncal fat was calculated as the average of axilla, chest, abdominal, and back measurements.
† WG ranges were used to be equivalent to BMI categories for women and men, respectively: lean, WG <77.5 and <89 cm; overweight, WG 77.5 to 88.2 cm and 89 to 101.1 cm; and obese, WG >88.2 and >101.1 cm.
‡ Significantly different from lean group (p <0.0001).
§ Significantly different from overweight group (p <0.0001).
∥ Percentage body fat was calculated from skinfold measurements using the Jackson-Pollock equation.
In women, plasma triglycerides, fasting glucose, systolic blood pressure, and non-HDL cholesterol increased progressively through each body-fat category ( Table 2 ). HDL cholesterol levels conversely decreased. ATP III MS prevalence in obese women was 58%, but only 30% had abridged ATP III MS. In overweight women, ATP III MS and abridged ATP III MS prevalence was much lower, only about 10% each. In men, as for women, plasma triglycerides, glucose, systolic blood pressure, and non-HDL cholesterol increased with increasing obesity, whereas HDL cholesterol levels decreased. In obese men, 74% had ATP III MS, whereas only 43% had abridged ATP III MS. In contrast, 20% of overweight men had ATP III MS and abridged ATP III MS. Finally, men had a higher percentage of current smoking than women.
