Metabolic Syndrome Risk Factors Revisited

Does the presence of metabolic syndrome make it easier to identifyindividuals at high risk for developing type 2 diabetes and cardiovasculardisease (CVD)? Or is it a less-than-accurate gauge that does little more thanmuddy the waters when contrasted with time-honored predictors?

Two recently published studies reveal markedly different viewpoints on thesubject.

Mercedes R. Carnethon, PhD, of the Feinberg School of Medicine,Northwestern University in Chicago, led the Coronary Artery Risk Developmentin Young Adults (CARDIA) study. Using the definition developed by the NationalCholesterol Education Program (NCEP), Carnethon and her colleagues set out toidentify modifiable risk factors for metabolic syndrome—a constellationof symptoms including insulin resistance, high blood pressure, and high lipidsestimated to affect 47 million adults in the United States. The studyconcluded that body mass index (BMI) and weight gain are important predictorsof the development of metabolicsyndrome.¹

The second study, the San Antonio Heart Study (SAHS), led by Michael P.Stern, PhD, of the University of Texas Health Science Center at San Antonio,sought to compare metabolic syndrome to the Diabetes Predicting Model and theFramingham Risk Score. The SAHS study concluded that metabolic syndrome, interms of reliability, is inferior to the two established predictingmodels.²

CARDIA

The CARDIA study concluded that a higher BMI in young adulthood and weightgain over 15 years are reliable predictors for development of metabolicsyndrome. In addition, the study offered evidence that younger adults whoconsume no alcohol and those who consume a higher proportion of their caloriesfrom carbohydrates and eat less crude fiber may be at increased risk overmoderate drinkers and those who get a smaller proportion of their caloriesfrom carbohydrates and eat more crude fiber.

The study encompassed 5,115 African-American and white men and women aged18–30 years at baseline. Participants were selected via random samplingfrom Chicago, Birmingham, Ala., and Minneapolis, Minn., as well as from theKaiser-Permanente health plan in Oakland, Calif. Participants were reexaminedsix times between baseline (1985–1986) and2000–2001.

After exclusions for missing measurements of metabolic syndrome componentsat baseline, pre-existing metabolic syndrome, pregnancy, and loss tofollow-up, 4,192 participants remained.

Collecting demographic characteristics (age, sex, and race), the studyassessed leisure time and work-related physical activities using a validatedquestionnaire. Participants whose physical activity was above the median atall four examinations were categorized as maintaining regular physicalactivity. Those whose activity fluctuated were categorized as moderatelyactive. Those whose activity fell below the baseline median at all fourexaminations were classified as engaging in low activity.

Metabolic syndrome was defined using NCEP guidelines as the presence of atleast three of the following:

1. Fasting glucose >110 mg/dl

2. Waist circumference >35 inches (women) or >40 inches (men)

3. Systolic blood pressure >130 mmHg or diastolic blood pressure >85mmHg

4. Triglycerides >150 mg/dl

5. HDL cholesterol <50 mg/dl (women) or <40 mg/dl (men)

SAHS

The SAHS study encompassed a population-based sample of 1,709 initiallynondiabetic participants, all of whom were followed for 7.5 years. One hundredninety-five of the participants developed type 2 diabetes. During the samelength of time, 156 of 2,570 participants experienced some kind ofcardiovascular event. The study also followed a population-based sample of1,353 initially nondiabetic Mexico City Diabetes Study (MCDS) participants for6.5 years. One hundred twenty-five members of the second group developed type2 diabetes.

In the SAHS dataset used to compute the incidence of diabetes, theprevalence of metabolic syndrome was 19.4% (16.0% in non-Hispanics and 21.1%in Mexican Americans). In the CVD dataset, the corrected prevalence ofmetabolic syndrome was 24.1% (16.8% in non-Hispanics and 27.6% in MexicanAmericans). “The higher prevalence in Mexican Americans is expected,since they are well known to have a greater frequency of many of the riskfactors that have been used to define metabolic syndrome,” saysStern.

DATA VALUE AND LIMITATIONS

According to Stern, other formulae with higher sensitivities and lowerfalse-positives do a better job than metabolic syndrome criteria of predictingwho will get diabetes or CVD.

It is possible that someone without metabolic syndrome still coulddemonstrate one or two of its components that put that person at risk, Sternpoints out. “By overemphasizing metabolic syndrome, there is a dangerthat physicians might overlook the importance of such factors as dyslipidemia,hypertension, etc., when they occur in isolation—for example, [in]patients who have abnormalities of one or more of these risk factors, but whodon't meet the criteria for metabolic syndrome.”

“[Stern] addresses an important issue that is oftenoverlooked—that we already know the primary risk factors for developingtype 2 diabetes and CVD,” says Carnethon. “I agree with Dr. Sternand his colleagues that metabolic syndrome adds little to already-establishedrisk factors for the disease. But I also agree with the limitations they statethat are inherent in a comparison of two very different models for comparingrisk.”

According to Carnethon, what is important and meaningful about metabolicsyndrome is that its components are subclinical cardiovascular disease factorsas opposed to clinical risk factors. Metabolic syndrome is composed of five ofthese types of “subclinical” (i.e., slightly elevated, but notelevated enough to warrant treatment) risk factors. “For example, highblood pressure [with] systolic blood pressure greater than 130 or diastolicblood pressure greater than 85 is a metabolic syndrome component rather thanhypertension [with] systolic blood pressure greater than 140 or diastolicblood pressure greater than 90,” she says. “The first bloodpressure values are not the clinical definition for hypertension that wouldwarrant drug therapy, whereas the second values are.”

Adds Carnethon, “This contribution of metabolic syndrome is that riskis elevated when multiple subclinical risk factors are present. In otherwords, elevated blood pressure [below the cutoff point for disease] alone maynot be as strong a risk factor as it would be if it were also combined withabdominal obesity or elevated fasting glucose or elevated blood lipids.”While it's true that having elevated blood pressure or a large waist alone maynot increase the risk for developing type 2 diabetes or CVD, the tollassociated with metabolic syndrome suggests that having a combination of riskfactors may increase risk within the population.

▪