Obesity

The world is getting fatter. If you're a Red Sox fan, as I am, and go to Fenway Park and sit in the cheaper seats that were built in 1912, they are only 18 inches wide. Not many of us today can fit into an 18-inch-wide seat. The box seats, which I measured last year, are 29 inches across; fortunately, I fit comfortably in those seats. In the apparel industry, in the last 8 or 9 years, the average neck size of a man has grown by an inch; and the average waist size has grown 4 inches. Obesity has become a major problem in the United States, and the increase in obesity precisely parallels the increase in the prevalence of diabetes that we have also seen in this country. Obesity is also a significant problem in children. If you see someone in his 20s today with diabetes, it's more likely that he (or she) has type 2 diabetes than type 1 diabetes, largely because he had impaired glucose tolerance in his teen and adolescent years.

Insulin Resistance and Visceral Fat

Insulin resistance is not an “all or none” or “yes/no” phenomenon—it is a continuum. There is also a continuum of risk associated with insulin resistance; and there is a more-or-less linear relationship of cardiovascular risk to insulin resistance, across the spectrum of patients who, at the low end, may be insulin-resistant with normal fasting glucose and normal postprandial glucose, to patients who have metabolic syndrome and impaired fasting glucose, to patients with overtly impaired glucose tolerance, and all the way up to patients with frank type 2 diabetes. Across this spectrum, insulin resistance appears to be mediated largely by visceral fat, which is very different from subcutaneous fat. There are very few National Football League players and sumo wrestlers who are insulin-resistant, even though they may be obese. Fat makes some good things—adiponectin, for instance, one of the “good” cytokines. But the minute they retire, NFL players and sumo wrestlers frequently become insulin-resistant and develop diabetes, because their fat moves from the subcutaneous compartment to the visceral compartment.

There's a lot of debate about how to measure visceral fat. Is it BMI? The ratio of visceral fat to subcutaneous fat? Waist circumference? Waist/hip ratio? All of these have their proponents, but in my view, probably the best is a true waist circumference (above the umbilicus) and a true hip circumference. This gives you a true ratio of visceral fat to fat in the remainder of the body.

Visceral Fat and Inflammation

There are 2 current theories on the relationship between visceral fat and inflammation. Visceral fat is a metabolically active organ. The adipocytes in visceral fat are actually different from the adipocytes in subcutaneous fat; there are many phenotypic similarities between the pre-adipocyte of visceral fat and macrophages. There are some who believe that macrophages in visceral fat may actually derive from pre-adipocytes. There are others who believe that pre-adipocytes or adipocytes release chemoattractant proteins that recruit macrophages into visceral fat, which themselves release more chemoattractants, creating a vicious cycle. Regardless of where they come from, the end result is that visceral fat is extensively infiltrated by a wide variety of inflammatory cells, which, in turn, make a whole host of cytokines (TNF-alpha, interleukin-6, interleukin-13, interleukin-18, etc). What can these visceral fat-derived cytokines do? TNF-alpha, infused into the forearm of volunteers, interferes with the local uptake of glucose, promotes local insulin resistance, and blocks the forearm's vasodilation capacity, thereby creating an area of local endothelial dysfunction. These cytokines can have profound effects on vascular function.

Fortunately for us, the body usually has counterbalancing forces; for every harmful pathway there is usually a beneficial pathway. Earlier, I mentioned adiponectin, one of the good cytokines, which is produced mostly by subcutaneous fat, not visceral fat. Higher levels of adiponectin actually increase your insulin sensitivity via the tyrosine receptor, and facilitate the transport of glucose into cells. Adiponectin also has a variety of other anti-inflammatory effects, and acts to decrease the levels of TNF-alpha, reduce CRP, decrease the number of circulating adhesion molecules, prevent vascular smooth-muscle-cell migration in areas of intimal injury, and decrease NF-kappa B (which regulates a number of genes related to atherothrombosis).

What about another form of fat—namely epicardial fat—frequently found in patients with coronary artery disease? Very interestingly, epicardial fat has the same characteristics as visceral fat. Both fats originate from brown adipose tissue. We now believe that there is actually a direct local effect of epicardial fat, as it releases harmful cytokines directly into the cardiac muscle, from the outside-in. If you biopsy epicardial fat at the time of CABG surgery, you will also find dense infiltration by a wide variety of inflammatory cells, including T-cells, mast cells, and macrophages. There is an approximately 10- to 100-fold increase of inflammatory cytokines in epicardial fat, compared with subcutaneous fat.

Furthermore, the accumulation of fat in the heart is a function of high levels of circulating free fatty acids. When you look at hearts taken from nonischemic cardiomyopathy patients who have undergone transplantation, lipid accumulation tends to occur in patients who are either obese or diabetic. This not only involves epicardial fat, but fat in the myocardial cells themselves, intramyocardial lipid overload, and an increase in the PPAR*-alpha activity. This increases the ability to oxidize free fatty acids, but it's still not adequate to meet the needs of the failing heart. It also causes a shift in myofibril expression from alpha to beta: alpha myofibrils display impaired ATP-ase activity that may result in contractile dysfunction. TNF-alpha has also been identified as a factor related to poor remodeling.

Metabolic Syndrome

Approximately 60% of people with metabolic syndrome have obesity as a contributing factor. Only about 10% to 20% of patients with metabolic syndrome actually have impaired fasting glucose. In the Kuopio study, the risk of cardiovascular death with metabolic syndrome was about 10% at 10 years, and the risk of death, nonfatal MI, or stroke was 18%. Despite this degree of risk, current guidelines contain no recommendation for statin therapy in metabolic syndrome patients with LDLs less than 130. A very interesting observation in the Kuopio study is that this increased risk for cardiovascular death was only modestly attenuated after correcting for most of the factors that actually define metabolic syndrome, such as low HDL, high triglycerides, high LDL, and obesity. It appears that it is the insulin-resistant component of the metabolic syndrome that is responsible for some of the excess mortality. This has prompted studies of the cardiovascular risk-reducing effects of drugs that specifically target insulin resistance: thiazolidinediones (TZDs) are the most common, but now also the PPAR-alpha and -gamma drugs and the pan-PPAR drugs are in development.

Another interesting aspect of metabolic syndrome is that there is independent risk associated with all of the individual qualifying criteria. Although you need 3 criteria to establish the diagnosis, even 1 or 2 criteria increase the risk for cardiovascular disease. If you have a high triglyceride and low HDL, you have almost all of the risks that you see in people who may have the full complement of 3 qualifying criteria, probably because it is a very good surrogate for insulin resistance. In the National Health and Nutrition Examination Survey (NHANES), the diagnosis of metabolic syndrome—even after adjusting for the risks associated with the individual criteria that define it—confers about a doubling of risk, which suggests that to treat these patients you may have to do more than raise HDL, lower triglyceride, lower blood pressure, or even get them to lose weight. These patients may derive additional benefits from treatments targeted specifically at insulin resistance.

One can view metabolic syndrome as being associated with end-organ disease. Metabolic syndrome patients consistently have thicker carotid intima and media, and stiffer vessels, than normal: this persists even after correction of the individual risk factors. Electron beam CTs in metabolic syndrome patients demonstrate higher calcium scores than age- and sex-matched controls, values that actually fall in the intermediate range between patients without metabolic syndrome and patients with overt diabetes. These observations also support the notion that insulin resistance has an important role in creating end-organ disease.

Data from Lahey Clinic, my own institution, show that 60% of 85 consecutive patients with acute MI who were under the age of 45 met the criteria for metabolic syndrome. Admittedly, 12 of these had undiagnosed diabetes, but in the literature from 15 to 20 years ago, it was either smoking or familial hypercholesterolemia that were the major risk factors for premature atherosclerosis. I would add that the Framingham risk score in our young patients with acute MIs was very, very low; Framingham risk score doesn't incorporate many of the criteria of metabolic syndrome.

Obesity

Weight alone is not the issue. Remember that metabolic syndrome can occur in people who are not overweight; in NHANES III, it was found that 5% to 10% of the population with BMIs in the 20 to 25 range may have metabolic syndrome. These people may not be obese, but they may have substantial visceral fat; or, in much thinner people without subcutaneous fat, they may not be making enough adiponectin. A little bit of subcutaneous fat may be viewed as a good thing. And there are probably populations of lean people who have small amounts of visceral fat that happens to be intensely metabolically active, which results in insulin resistance and metabolic syndrome. There are plenty of people who are very thin, but who are very insulin-resistant, and people who are very fat, but very insulin-sensitive. It's not just the weight.

But what about losing weight? You can lower your CRP dramatically if you lose 20 to 30 pounds. In most people who are really obese, however, weight loss may lower CRP, but not to lower-risk levels (<2), since CRP was probably substantially elevated to start with (frequently in the 5–8 range). In obese patients with metabolic syndrome, weight loss and exercise may be the safest way to lower CRP, lower blood pressure, and improve the lipid profile, but often it is not enough to affect insulin resistance.

Risk Stratification and Treatment

How do I suggest treating the tens of millions of metabolic syndrome patients, particularly younger patients? Do we treat a 39-year-old patient with metabolic syndrome the same way as a 60-year-old? There are a couple of higher-risk cohorts that can be identified. A female with metabolic syndrome at an early age is a pretty high-risk individual, almost as though she has early diabetes. For any number of components of the metabolic syndrome, a woman will have twice the CRP level that a man will. Also, if you have a strong family history of premature cardiovascular disease and metabolic syndrome, that, for me, is a CHD equivalent, similar to overt diabetes. I would strongly consider using a statin in such a person, even though the LDL may be 100 and the patient is 45 years old. If a patient has a 1st-degree relative with type 2 diabetes, he or she has at least a higher risk for conversion to type 2 diabetes. A patient with other underlying inflammatory disease (and presumably high CRP levels) and metabolic syndrome is also at higher risk for cardiovascular events. There are so many young people getting imaging studies these days that we are seeing a lot more subclinical vascular disease. I believe that if a patient, even a young patient, has documented vascular disease of any kind and metabolic syndrome, this merits diabetes-like CHD-equivalent prevention with aspirin and statins. And, as I mentioned before, high triglyceride/low HDL patients are more likely to be insulin-resistant. And since insulin resistance carries the greatest degree of overall risk, these may be particularly at-risk individuals.

What else do you do for higher-risk patients? I may start metformin in some of my metabolic syndrome patients who have impaired fasting glucose, because they are also very likely to have impaired glucose tolerance; several studies have shown that metformin may help prevent conversion to diabetes in patients with impaired glucose tolerance. There may even be a role for low-dose TZDs in patients who are insulin-resistant with at-risk features, because those drugs are targeted toward insulin resistance. However, this is still theoretical; we need actual clinical trials to see if that strategy decreases cardiovascular risk.