Department of Medicine, St. Luke's/Roosevelt Hospital Center, 1111 Amsterdam Avenue, Room 1020, New York, NY 10025. E-mail: firstname.lastname@example.org
The metabolic syndrome has been characterized by a cluster of abnormalities that include obesity, hyperglycemia, dyslipidemia, and hypertension. Other conditions associated with this syndrome include microalbuminuria, inflammation, a prothrombotic state, and a fatty liver. Together, these abnormalities lead to an environment where the risk of developing both type 2 diabetes and atherosclerotic cardiovascular disease are greatly enhanced. Recognition of this syndrome by practitioners, early treatment, and long-term management are crucial for disease prevention. Successful treatment requires the introduction of lifestyle changes initially and pharmacotherapy subsequently if lifestyle changes are not sufficient.
The metabolic syndrome has been defined by two different groups in a somewhat different manner. This has led to some confusion and controversy. The Adult Treatment Panel III (ATP III)1 has defined metabolic syndrome as a state in which three of the six characteristics (abdominal obesity, atherogenic dyslipidemia, elevated blood pressure, insulin resistance and/or glucose intolerance, prothrombic state, proinflammatory state) listed in Table 1 are present (1). The definition set forth by the World Health Organization (WHO) (Table 2) includes diabetes or impaired glucose tolerance (IGT) and also lists thresholds for the risk factors of insulin resistance, raised arterial pressure, raised plasma triglycerides, central obesity, and microalbuminuria (2).
Table 1. . ATP III definition of the metabolic syndrome*
Risk Factors for Metabolic Syndrome
Defined as a state in which three of the six characteristics in the table are present (1).
Raised blood pressure
Insulin resistance and/or glucose intolerance
Table 2. . WHO definition of the metabolic syndrome*
Defined as impaired glucose tolerance or type 2 diabetes and/or insulin resistance plus two or more of the risk factors listed (2).
Impaired glucose regulation or diabetes
1. Type 2 diabetes
2. Impaired glucose tolerance
3. Impaired fasting glucose
Under hyperinsulinemic, euglycemic conditions, glucose uptake below lowest quartile for background population under investigation
Women, waist-to-hip ratio > 0.85, and/or BMI > 30 kg/m2
Urinary albumin excretion rate ≥ 20 μg/min or albumin-to-creatinine ratio ≥ 30 mg/g
It is clear that using different criteria such as impaired fasting glucose (IFG) instead of IGT will identify different subsets of patients (3, 4). In addition, the inclusion of frank diabetes as a criterion in the WHO definition adds an entire group with more impairment of insulin action and a greater risk for coronary heart disease (CHD) (5). Defining overweight based on waist circumference rather than BMI is not likely to matter when BMI is > 35; however, at a lower BMI cut-off, more men than women are likely to be represented.
Both the ATP III and WHO definitions were developed to identify a group of risk factors that would have a higher probability of predicting subsequent development of CHD. The goal of the ATP III guidelines was to alert practicing physicians regarding the cluster of risk factors, with a 2-fold objective: to reduce underlying causes for the risks (i.e., obesity and physical inactivity) and to treat associated lipid and nonlipid risk factors (1). For the purpose of this short review, the ATP III definition will be used.
The concept of insulin resistance as underlying a cluster of risk factors or end points of disease originated with Dr. Gerald Reaven in 1988 (3). Insulin resistance syndrome, or syndrome X, as he called it, was originally described in lean individuals. By measuring total body glucose use at a given level of insulin, he depicted a population of lean, normal volunteers with a wide range of sensitivity to the same level of insulin. He arbitrarily took the lower 25% of the sample and labeled them insulin-resistant.
It soon became clear that an individual could be insulin-resistant from one of two main reasons: he/she could be genetically resistant (like Reaven's group) or could acquire the resistance by becoming obese (4). The impact of obesity, independent of genetic factors, was illustrated in a study of 23 sets of identical twins who were discordant for weight (6). Within a twin pair, the obese member had higher insulin levels and lower insulin sensitivity, and these differences were particularly evident among obese members with high abdominal fat distribution. In the modern world, it is obesity that tends to drive the development of insulin resistance and, therefore, predisposes a person to a much higher risk of developing type 2 diabetes and CHD (7).
Once hyperinsulinemia and insulin resistance are present, a cascade of metabolic changes occurs that leads to dyslipidemia, hypertension, hyperglycemia, and eventually type 2 diabetes and CHD. Individuals who develop insulin resistance, due to either their genetic make-up or increased obesity, also develop specific abnormalities of glucose metabolism, fatty acid metabolism, vascular reactivity, inflammatory responses, and coagulation defects. In addition, these people are more prone to specific clinical disorders, such as polycystic ovarian syndrome, nonalcoholic fatty liver disease, hyperuricemia, and gout (8).
Mechanism of Impaired Insulin Action
Both in vivo and in vitro studies have identified mechanisms underlying the defects in insulin action. Bergman et al. (9) have shown the relationship between insulin sensitivity and glucose levels by using a mathematical model that is based on the disappearance of injected glucose over time. Quantification of the insulin index in 93 healthy volunteers showed that insulin sensitivity varies widely with BMI (10). The lower insulin sensitivity in the obese is also seen when insulin levels are monitored over a 24-hour period (11). In both the fasting and postprandial states, obese subjects require insulin levels that are several times higher than nonobese subjects to maintain normal glucose levels (11).
Under normal physiological conditions, insulin binds to its receptor at the cell surface, leading to tyrosine autophosphorylation and consequent intracellular signaling. This pathway includes translocation of glucose transporter to the cell surface, which allows glucose to enter the cell. Once inside the cell, glucose is oxidized or stored as glycogen for later use. In contrast, in insulin resistance, the insulin receptor activity is impaired. It has been shown that in obesity, protein kinase activity of the insulin receptor is defective (12). Thus, the initial step that will allow appropriate phosphorylation of the receptor and initiate the internal signals for adequate insulin action is impaired.
The Role of Free Fatty Acids (FFAs)
FFAs are an alternate fuel to glucose in many tissues of the body. As such, the use of glucose is closely connected with the use of FFAs, which are released from adipose cell depots through the action of hormone-sensitive lipase. Lipase, in turn, is activated when the activity of insulin is low, as occurs in insulin resistance. The more insulin-resistant the patient, the greater the impairment of insulin action of fat cells, leading to increased lipolysis and plasma FFA elevation (13, 14). An increased level of FFA has profound effects on liver and skeletal muscle (15). In liver, it enhances FFA oxidation, which is a powerful stimulus to gluconeogenesis (15). In skeletal muscle, glucose use is inhibited as FFA oxidation increases with the higher FFA levels (16). Boden and others (17, 18) have shown that elevating FFA impairs total body glucose use. On the other hand, Santomauro et al. (19) have shown the opposite, that by lowering circulating FFA with acipimox, glucose use can be increased.
The effect of insulin resistance in muscle leads to impairment of both glucose oxidation and glucose storage but has a greater effect on glucose storage (20). Recent investigative breakthroughs using nuclear magnetic resonance spectroscopy have suggested that the principal effect is at the glucose transport step, which involves blockage of PI3 kinase, an enzyme that mediates glucose transporter translocation in the muscle cell (21). A proposed model for this mechanism suggests that the intracellular fatty acid metabolites (possibly diacylglycerol, fatty acyl CoAs, or ceramides) activate a serine/threonine kinase cascade leading to phosphorylation of serine/threonine sites on insulin receptors. The serine phosphorylated sites fail to activate PI3 kinase as would occur if threonine sites were phosphorylated, resulting in decreased glucose transport and other downstream events (22). Boden and Shulman (23) have proposed that the decrease in insulin receptor substrate 1 tyrosine phosphorylation suppresses PI3 kinase activity and decreases glut 4 translocation, leading to decreased glucose transport.
In parallel to the decreased glucose transport due to increased FFA, there is enhanced gluconeogenesis and hepatic glucose production (24, 25). Thus, when the skeletal muscle is taking up less glucose, the liver is producing more glucose, leading to increased glucose levels, which stimulates increased insulin secretion from β cells. In individuals who are genetically predisposed to diabetes, the continued stress on higher insulin needs leads to eventual β cell exhaustion and the development of type 2 diabetes (26). In part, the eventual apoptosis of β cells may be related to FFAs or FFA metabolites in the β cell itself. A suggested mechanism for β cell death is that surplus FFA or acyl-CoA can increase nitrous oxide generation, leading to cell dysfunction and eventual cell death (27).
Role of Adipocyte Hormones
Adipocytes are endocrine cells that release many circulating hormones and metabolites that have physiological actions on distant tissues. Examples of factors released that are likely to be important in insulin resistance include adiponectin, resistin, leptin, tumor necrosis factor α, and interleukin-6. Adiponectin, which is produced only by white adipose tissue (28), is reduced in states of insulin resistance, such as obesity and type 2 diabetes (29), and increases insulin sensitivity (30). In the muscle, it appears to activate adenosine monophosphate-activated protein kinase (31). A second protein, resistin, which is also secreted from adipose tissue, has been shown to enhance insulin resistance in rodents (32, 33), although its effect in humans is not yet clear.
Consequences of the Insulin-Resistant State
As discussed below, there are many consequences from risk factors associated with the insulin-resistant state or metabolic syndrome. The primary risk factors associated with metabolic syndrome include dyslipidemia, high blood pressure, elevated glucose, and a large waist circumference. Dyslipidemia can be characterized further by the presence of low high-density lipoprotein-cholesterol (HDL-C) and elevated triglycerides. Although the low-density lipoprotein cholesterol (LDL-C) may be normal or only slightly elevated, the LDL particles are smaller, denser, and more atherogenic (34). In people with insulin resistance, a small absolute change in blood pressure can translate into a hypertension prevalence of ∼20% in an age-adjusted population (35). In addition to the blood glucose elevation discussed previously, a proinflammatory state is present in a majority of these individuals, which is manifested by high levels of C-reactive protein, interleukin-6, and other cytokines (36, 37, 38).
Ferrannini et al. (35) have analyzed the relationship between insulin sensitivity and blood pressure and have shown that insulin sensitivity contributes to blood pressure variability. They found that for each 10-unit increase in insulin resistance (10 μM/min per kilogram decrement in molar value), there was an increase of systolic blood pressure of 1.7 mm Hg and an increase of diastolic blood pressure of 2.3 mm Hg. It is known that an increase of 2 mm Hg predicts an increased risk of stroke by 17% and coronary artery disease (CAD) by 10% in the general population (39), and this risk is likely to be higher in patients with the metabolic syndrome. Possible mechanisms by which insulin resistance and hyperinsulinemia contribute to the pathogenesis of hypertension include: activation of the sympathetic nervous system, increased activity of the Na/H exchange pump, increased retention of renal sodium, and increased salt sensitivity. These mechanisms are not well understood, and it is not clear how important insulin resistance is to their development.
In addition to these factors, coagulation abnormalities are evident in insulin-resistant persons. In these cases, fibrinogen plasminogen activator inhibitor-1 (PAI-1), an important regulator of fibrinolysis (Figure 1) (40), is elevated (38); its elevation decreases the fibrinolytic activity, thus enhancing the chance for clot formation. Platelet function is also abnormal, with increased aggregation propensity and thrombin generation. A number of abnormalities of vascular function are associated with insulin resistance and the metabolic syndrome. Although the mechanisms responsible for this are not well understood, they may include resistance to insulin-mediated vasodilation, abnormal endothelial signaling, and increased sympathetic nervous system activity. In addition, a reduced release and responsiveness to nitric oxide, a potent vasodilator derived from endothelial cells, appears to be associated with this mechanism.
It is likely that the microalbuminuria that is often seen in patients with the metabolic syndrome is related to endothelial dysfunction because it affects the kidney. Microalbuminuria has been found to be a good predictor of subsequent CAD, which may be due to the fact that it reflects the abnormal vascular/coagulation/inflammatory state in the kidney.
The liver is also affected in the metabolic syndrome. Individuals with insulin resistance and dyslipidemia have a much higher incidence of fatty liver, also known as nonalcoholic steatohepatitis (NASH) (41, 42, 43). This condition can lead to inflammation, cirrhosis, and end-stage liver disease (44, 45). It is not clear whether insulin resistance is a cause of NASH or whether NASH leads to higher insulin resistance. This needs to be further explored.
Management of the Metabolic Syndrome
How should the metabolic syndrome be managed? There are two important approaches, both highlighted by the ATP III report. The first is to try to reduce the insulin resistance per se, and the second is to directly treat all of the specific risk factors that are abnormal.
Reducing insulin resistance requires lifestyle changes. Both a decrease in weight and an increase in activity can improve insulin resistance (46). Therefore, it is important that physicians encourage their patients to make an effort to incorporate both changes in their lifestyle. Both the Diabetes Prevention Program (47) and the Finnish Diabetes Prevention Trial (48) have shown convincingly that a modest decrease in weight and a modest increase in physical activity can prevent the progression of IGT to diabetes in a large number of cases. Dr. Wadden has discussed lifestyle change in another article in this supplement, so no further elaboration is required, except to restate how important this effort is in the overall management of the metabolic syndrome. If lifestyle changes are not successful in reducing weight in those individuals who are overweight or obese, it is reasonable to try pharmacotherapy to achieve greater success. At present, two drugs—orlistat and sibutramine—are approved for long-term use in the United States. These have been discussed by Dr. Klein in a preceding article.
The second goal is to treat abnormal risk factors. As emphasized by the ATP III report, the primary intervention is to focus on treating a risk factor that is not itemized in the metabolic syndrome definition. LDL-C continues to be a primary risk factor for CAD. Individuals with high LDL-C need to be treated initially with a diet (deficit in calories if overweight, low in total cholesterol, low in saturated and trans fats) to achieve a goal of LDL-C < 100 mg/dL. If this cannot be achieved by lifestyle modification alone, then pharmacological intervention with statins and bile acid sequestrants may be necessary. Large clinical trials have shown that both treatments can be effective (49, 50).
The second risk factor to treat would be elevated triglycerides. The treatment goal should be to reduce triglycerides to < 150 mg/dL, which can be achieved by diet alone. A diet in which 30% of calories are derived from fat, 15% from protein, and 55% from carbohydrates is acceptable if the diet is hypocaloric and the patient is losing weight. If the patient is weight-stable and overweight, it may be more appropriate to decrease carbohydrates and replace them with monounsaturated fats, which do not raise triglyceride levels (51). Finally, adding fibrates may be necessary for desired reduction in triglycerides. Adding nicotinic acid is not appropriate because it may enhance insulin resistance.
It is more difficult to raise the level of HDL-C. Nonpharmacological intervention—by lifestyle changes that involve weight loss and increasing physical activity—is the best approach. Additionally, moderate alcohol intake may help, but only if substituted for other calories. Some drugs, such as β-blockers, anabolic steroids, and progestational agents, also may increase HDL-C levels.
In addition to lipids, a critical risk factor that should be controlled is blood sugar because type 2 diabetes increases the risk of CAD 2- to 4-fold (52, 53). The risk of myocardial infarction (MI) in patients with diabetes and no history of MI is as high as in patients without diabetes who have had a previous MI (54). Moreover, mortality after the first MI is higher in both men and women with diabetes than in their nondiabetic counterparts (55). If a patient has frank diabetes, it should be treated vigorously to reach an acceptable goal of glycosylated hemoglobin of < 6.5%. If lifestyle changes are not enough, then appropriate antidiabetic medication should be initiated. In subjects with IGT or IFG, a lifestyle change of lowering body weight and increasing physical fitness should be stressed. Such change has been shown to be a powerful weapon for the prevention of progression to diabetes. In addition to treating hyperglycemia, aggressive treatment of dyslipidemia and hypertension in patients with diabetes is extremely important (56, 57, 58).
High blood pressure in individuals with the metabolic syndrome should be a serious concern. Appropriate lowering of salt intake and high sodium foods should be initiated. If this is not adequate, a diet such as that described in the DASH (59) and Premier (60) trials should be instituted. These diets are low in cholesterol and saturated fats and high in fruits, vegetables, grains, and low-fat dairy products. If this still does not lower the blood pressure below 130/80 mm Hg, appropriate pharmacotherapy must be initiated. It is important to keep in mind the effect of some antihypertensive agents on insulin resistance. For instance, β-blockers and thiazide diuretics tend to decrease insulin sensitivity, whereas angiotensin-converting enzyme inhibitors and α-blockers tend to improve sensitivity (61). Calcium channel and angiotensin-II receptor blockers have no significant effect on insulin resistance.
Pharmacotherapy Targeted to Insulin Resistance
At present, there are two classes of drugs that target insulin resistance directly, the biguanides and the thiazolidinediones (TZDs). Metformin, the only biguanide approved for use in the United States, has been successful in both the Diabetes Prevention Program (47) and the Finnish Diabetes Prevention Trial (48) in decreasing the progression to type 2 diabetes in patients with IGT. It is not clear whether this decrease is related to improved insulin resistance or to another mechanism. The TZDs rosiglitazone and pioglitazone improve insulin sensitivity, endothelial function, markers of vascular inflammation (62), and carotid intima medial thickening (63), and inhibit vascular smooth muscle growth and migration. The only randomized clinical trial demonstrating a favorable effect of a now discontinued TZD, troglitazone, on preventing conversion to diabetes (64) was in women with a diagnosis of gestational diabetes.
Emerging evidence indicates that diabetes and CAD share many metabolically linked common precursors and that diabetes is a major risk factor for CAD (65). Long-term trials for prevention of type 2 diabetes and CAD are clearly necessary.