Epidemic Obesity and the Metabolic Syndrome

Obesity, type 2 diabetes, and the metabolic syndrome are multifactorial diseases of considerable heterogeneity.⁸ However, whereas diagnostic criteria for obesity and for type 2 diabetes are clear cut, this is not the case for the metabolic syndrome. There have been a number of attempts to develop standardized criteria for the diagnosis of the metabolic syndrome. The World Health Organization (WHO) developed a definition in 1998 that stated that individuals need to show evidence of insulin resistance and at least 2 of 4 other factors (hypertension, hyperlipidemia, obesity, and microalbuminuria).⁹ Isomaa et al¹⁰ suggested that this definition of the metabolic syndrome strongly predicted cardiovascular disease in the Botnia, Finland, population. A more recent version of the WHO definition¹¹ (Table 1) has been described using a lower cutoff for hypertension, 140/90 versus 160/90 mm Hg.⁹ In 2001, the National Cholesterol Education Program (NCEP) suggested another definition for the metabolic syndrome (Table 2),¹² which required at least 3 of 5 factors to be present for definition of the metabolic syndrome. The 5 factors are the following: increased waist circumference, hypertriglyceridemia, low HDL cholesterol, hypertension, and a fasting glucose of 110 mg/dL or higher. This definition is easier to use in clinical practice because glucose tolerance testing, insulin concentration measurements, and microalbuminuria testing are not required. Lemieux et al¹³ have introduced an even simpler definition of the metabolic syndrome in men, “the hypertriglyceridemic waist” (Table 3). Because of its greater clinical applicability, this review will emphasize the NCEP definition of the metabolic syndrome.

Recently, the Third National Health and Nutrition Examination Survey (NHANES) reported on the prevalence of the NCEP-defined metabolic syndrome.¹⁴ The overall prevalence of the metabolic syndrome in adults over the age of 20 years was 24%, but the age-specific rate increased rapidly. The prevalence in 50-year-old subjects was >30%, and the prevalence in subjects age 60 years and over was 40%. In addition, the prevalence was highest in Hispanics and lower in non-Hispanic whites and in African Americans. The lower prevalence among African Americans may be explained by the 2 separate lipid criteria defined by the NCEP (high triglycerides and low HDL cholesterol), which offset the higher rates of hypertension and glucose intolerance observed in this ethnic group.

The prevalence of coronary heart disease (CHD) in the NHANES population over the age of 50 has recently been explored by Alexander et al.¹⁵ In this study, the prevalence of the NCEP-defined metabolic syndrome among diabetic subjects was 86%. A lower (but still higher-than-average) prevalence of the metabolic syndrome was observed in subjects with impaired glucose tolerance (31%) and impaired fasting glucose (71%). The prevalence of the metabolic syndrome in the NHANES study was 60% greater than the prevalence of type 2 diabetes in the same population. In addition, the prevalence of CHD in nondiabetic subjects who also had the metabolic syndrome was intermediate, falling between the prevalence among nondiabetic subjects without the metabolic syndrome and diabetic subjects with the metabolic syndrome (Figure). Interestingly, the relatively rare diabetic subjects without the metabolic syndrome (≈15%) had a prevalence of CHD similar to that of nondiabetic subjects without the metabolic syndrome. Although these results need to be replicated in other populations, and particularly in prospective studies, these observations suggest that subjects with the NCEP-defined metabolic syndrome have an intermediate risk of CHD and are not equivalent in risk to subjects with only CHD or type 2 diabetes.

Abnormalities of glucose, insulin, and lipoprotein metabolism are common in patients with hypertension,^16,17 and hyperinsulinemia has been proposed as the link between hypertension, obesity, and impaired glucose tolerance.¹⁸ The mechanism behind this association is unclear, as short-term insulin infusion induces skeletal muscle vasodilation¹⁹ that is mediated by NO²⁰ and results in a decrease in systemic vascular resistance.²¹ Because insulin is a direct vasodilator, other physiological mechanisms will have to come into play if insulin is to have a causal role in the pathogenesis of hypertension.¹⁷ In normal individuals, acute increases in plasma insulin within a physiological range increase sympathetic neural outflow without elevating arterial pressure.²² It is reasonable to postulate that the sympathetic nervous system overrides the normal vasodilatory effects of insulin under more extreme conditions such as sucrose feeding,²³ in obesity, and with hypertension.²⁴ Even lean individuals with essential hypertension have insulin resistance and hyperinsulinemia.^16,25 In short, the link between insulin resistance and hypertension is given through the sympathetic nervous system.

Obesity, type 2 diabetes, and insulin resistance are also important risk factors for the development of heart failure.^26,27 Conversely, heart failure causes insulin resistance and is associated with increased risk for the development of type 2 diabetes.²⁸ Like the development of cardiovascular disease due to impaired intracellular insulin signaling,²⁹ the development of insulin resistance with heart failure is likely to be multifactorial.²⁷ Heart failure may cause insulin resistance by sympathetic overactivity, impaired endothelial function, loss of skeletal muscle mass, or increased circulating cytokines such as tumor necrosis factor α. We have argued that a vicious cycle is set into motion in which heart failure and insulin resistance worsen each other.²⁷

Recently, much attention has been given to the metabolic syndrome. Ridker et al³⁰ have shown that C-reactive protein (CRP), a marker of subclinical inflammation, strongly predicts the risk of coronary events. In addition, CRP predicts the development of coronary events even after Framingham global risk is considered.³¹ One link between subclinical inflammation and CHD may be the metabolic syndrome and insulin resistance. In nondiabetic IRAS subjects, CRP levels were significantly correlated with cardiovascular risk factors (correlation of CRP with body mass index, 0.40; with waist, 0.43; with systolic blood pressure, 0.20; with fasting glucose, 0.18; with fasting insulin, 0.33; and with insulin sensitivity, −0.37; all probability values, <0.001).³¹ In addition, the level of CRP was strongly correlated with the number of metabolic disorders (dyslipidemia, upper body adiposity, insulin resistance, and hypertension).³¹ Consistent with the observational studies discussed above, insulin-sensitizing pharmacological agents such as rosiglitazone reduced CRP levels by ≈25% in diabetic subjects.³² This result is similar to the effect of various statins in reducing CRP in other studies.³³

The NCEP¹² suggests that behavioral interventions promoting weight loss and increasing physical activity are the basis of therapy for the metabolic syndrome (Table 4). Indeed, weight loss and increased physical activity have been shown by the Diabetes Prevention Program to reduce the risk of type 2 diabetes by 58%.³⁴ This reduction was greater than that seen with metformin in subjects with impaired glucose tolerance (25%). Additionally, the NCEP suggests that subjects with the metabolic syndrome be treated for underlying conditions such as hypertension, diabetes, and lipid disorders. Currently, it is controversial whether nondiabetic subjects with the metabolic syndrome should be treated with insulin-sensitizing therapies. A case could be made for such intervention given that thiazolidinediones can reduce CRP to the degree that has been observed with statin therapy.^32,33 However, insulin-sensitizing therapies have not yet been shown to reduce cardiovascular disease in randomized clinical trials.

The NCEP does not specify whether subjects with the metabolic syndrome should receive more intense therapy for underlying conditions (ie, hypertension, lipid disorders) than that called for by their estimated global risk based on the Framingham Study. A possible approach to this issue is given in Table 5. Global risk should be calculated for all subjects with the metabolic syndrome, even if they have <2 major risk factors. For example, if a subject has a global risk of 5% to 10% (corresponding to a LDL cholesterol goal of <160 mg/dL) and also has the metabolic syndrome, one might consider treating this subject as if he/she had a global risk of 10% to 20% (using an LDL cholesterol goal of <130 mg/dL).

If diabetes is considered a model for the metabolic syndrome and 85% of diabetic subjects have the metabolic syndrome, then a strong case can be made for drug treatment of the underlying conditions. Antihypertensive therapy using various initial treatments has been shown to be at least as effective in reducing cardiovascular morbidity and mortality in diabetic subjects as in nondiabetic subjects in the Hypertension Optimal Treatment (HOT) trial.³⁵ In addition, statin therapy was shown to reduce coronary events in diabetic subjects in the Scandinavian Simvastatin Survival Study (4S).³⁶ However, most subjects with the metabolic syndrome do not have diabetes.¹⁵ Unfortunately, few data exist on the treatment of nondiabetic subjects with the metabolic syndrome. In the 4S study, statin therapy reduced the prevalence of CHD in subjects with diabetes and in subjects with impaired fasting glucose.³⁶ Additionally, statin therapy was more effective in the 4S subjects with the lipid triad (high LDL cholesterol, high triglycerides, and low HDL cholesterol) than in subjects with isolated high LDL cholesterol.³⁷ Although the latter 2 reports from the 4S study did not test a specific definition of the metabolic syndrome, they imply that subjects with characteristics of the metabolic syndrome are likely to benefit from statin therapy.

An increasingly recognized risk factor for cardiovascular disease is the metabolic syndrome. Although many investigators believe that insulin resistance is the underlying cause of it, some features, such as hypertension, are more weakly correlated with insulin resistance. The NCEP definition of the metabolic syndrome is relatively simple and is related to risk of cardiovascular disease. Primary treatment should be behavioral intervention, but treatment of existing comorbid conditions such as hypertension and dyslipidemia needs to be considered. It is likely that subjects with the metabolic syndrome will receive more aggressive therapy than those with similar global risk without the metabolic syndrome. However, no guidelines address this issue at present.

The opinions expressed in this article are not necessarily those of the editors or of the American Heart Association.