- Top of page
Prev Cardiol. ****;**:**–**.
It has been suggested that within the traditional body mass index (BMI) categories there is a heterogeneous pattern of cardiometabolic risk factor clustering. The objective of this research was to determine the associations among obesity, cardiometabolic abnormalities, and cardiovascular disease (CVD) in a large population-based study of Appalachian adults. The study comprised a cross-sectional survey of Appalachian adults residing in 6 communities in Ohio and West Virginia, who were aged 18 years and older (n=14,783, 50.9% women). The authors categorized BMI into normal weight (<25 kg/m2), overweight (25–29.9 kg/m2), and obese (≥30 kg/m2). Cardiometabolic abnormalities were defined as the presence of hypertension, elevated triglycerides (≥150 mg/dL), decreased high-density lipoprotein cholesterol (<40 mg/dL [men], <50 mg/dL [women]), elevated fasting glucose (≥100 mg/dL)/diabetes, insulin resistance (homeostasis model assessment >5.13), or elevated C-reactive protein (>3 mg/L). They found that 25.6% of normal-weight adults displayed clustering of ≥2 cardiometabolic abnormalities; in contrast, 36.8% of overweight/obese adults displayed no clustering. Compared with normal-weight persons without clustering of cardiometabolic abnormalities (referent), the odds ratio of CVD was 1.06 (95% confidence interval [CI], 0.84–1.34) among overweight/obese individuals without cardiometabolic clustering, 2.21 (95% CI, 1.74–2.81) among normal-weight individuals with cardiometabolic clustering, and 2.45 (95% CI, 2.02–2.97) among overweight/obese individuals with cardiometabolic clustering. These results suggest that within the traditional BMI categories, there may be heterogeneity of CVD risk depending on whether there is underlying clustering of cardiometabolic abnormalities.
Obesity has become a significant public health concern. According to the Behavioral Risk Factor Surveillance System (BRFSS), the prevalence of obesity in the United States is 26.6%.1 Obesity is an even bigger problem in the Appalachian region, where obesity prevalence rates are substantially higher than in the rest of the United States.2 Obesity has been linked to an increased risk of many disease states, including hypertension, diabetes, hypercholesterolemia, and cardiovascular disease (CVD).3
Recent studies have shown that clustering of metabolic risk factors such as high lipid levels, elevated fasting glucose, insulin resistance, and high C-reactive protein (CRP) are independently associated with the risk of CVD.4 However, the interrelationship between clustering of these cardiometabolic risk factors and obesity is not clear. In a recently published study, Wildman and colleagues5 demonstrated that a considerable proportion of normal-weight individuals (23.5%) displayed clustering of cardiometabolic risk factors, while a higher proportion of overweight (51.3%) and obese (31.7%) individuals did not display clustering of these factors. It is unclear whether this is an isolated finding.
In this context, we examined the hypothesis that a substantial proportion of Appalachian adults in the normal-weight body mass index (BMI) category demonstrate cardiometabolic risk factor clustering, while a substantial proportion of overweight/obese individuals display no clustering of cardiometabolic risk factors. We also examined the effect of joint exposure to overweight/obese BMI status and clustering of cardiometabolic abnormalities on the risk of self-reported CVD.
- Top of page
The prevalence rate of obesity in a representative sample of Appalachian adults (34.8%) was higher than the US national average (26.6%).1 We found that one-fourth of normal weight adults displayed cardiometabolic risk factor clustering, while more than one-third of overweight/obese adults did not display clustering of cardiometabolic risk factors. Older age, male sex, cigarette smoking, never drinking alcohol, and physical inactivity were the factors that were found to be associated with the presence of cardiometabolic risk factor clustering (ie, the metabolically abnormal phenotype) in normal-weight adults. Younger age, higher levels of education, never smoking, daily consumption of alcohol, and regular exercise were the factors that were associated with the absence of cardiometabolic risk factor clustering (ie, the metabolically healthy phenotype) in overweight/obese adults. When examining normal-weight persons without cardiometabolic risk factor clustering, we found that overweight/obese individuals without cardiometabolic risk factor clustering may not have a higher CVD risk; in contrast, normal-weight individuals who displayed cardiometabolic clustering had twice the risk of CVD. These findings suggest a complex interaction between cardiometabolic risk factors and obesity, bringing to light the need to intervene on these metabolic risk factors in addition to weight status.
Wildman and colleagues5 found that about one-fourth of normal-weight US adults displayed cardiometabolic risk factor clustering, while about one-third of obese adults did not; our results are similar to these estimates and extend these to an Appalachian population. Also, similarly to our study, they found that older age, male sex, and physical inactivity were associated with cardiometabolic abnormality in normal-weight individuals. In addition to confirming the findings by Wildman and colleagues, our study showed that overweight/obesity did not increase CVD risk in the absence of cardiometabolic risk factor clustering.
Previous prospective studies have only examined individual metabolic abnormalities such as abnormal lipid levels,8 insulin resistance,9 CRP,10,11 diabetes mellitus/hyperglycemia,12 or hypertension13 and provided stratified analysis by BMI. To our knowledge, previous prospective studies have not examined the risk of CVD or cardiovascular mortality comparing normal-weight persons with clustering of cardiometabolic risk factors vs overweight/obese persons without such clustering. Therefore, an important contribution of the current study is that we found that—compared with normal-weight individuals without cardiometabolic risk factor clustering—overweight/obese individuals without cardiometabolic risk factor clustering may not have a higher CVD risk, whereas normal-weight persons who displayed clustering of cardiometabolic risk factors had more than twice the risk of CVD.
Our sample was representative of a typical Appalachian population, with relatively higher rates of obesity, smoking, and physical inactivity. Modifiable lifestyle factors, such as smoking, physical inactivity, and abstinence from alcohol, were associated with cardiometabolic risk factor clustering (ie, the metabolically abnormal phenotype) in normal-weight adults. Cigarette smoking has been shown to cause lipid abnormalities, such as increased total cholesterol, low-density lipoprotein cholesterol, and triglycerides and decreased HDL cholesterol.14 Smoking also appears to increase risk of type II diabetes15 and cause elevations in inflammatory markers, such as CRP.16 Similarly, among normal-weight individuals in our study, former and current smokers were more likely to display cardiometabolic risk factor clustering.
Physical inactivity was associated with cardiometabolic risk factor clustering (ie, the metabolically abnormal phenotype) in normal-weight adults, while exercising at least once per week was associated with the absence of cardiometabolic risk factor clustering (ie, the metabolically healthy phenotype) in overweight/obese adults. A 12-week prospective study in obese adults with the metabolic syndrome showed significant reductions in cardiometabolic risk factors, such as blood pressure, insulin resistance, and lipid profile, with exercise alone.17 Addition of dietary restrictions improved weight loss but did not result in greater improvements of cardiometabolic risk factors, suggesting that the effect of physical activity on metabolic profile was independent of dietary changes.
Our study demonstrated a protective effect of alcohol consumption among obese adults, regardless of the amount of alcohol intake. Other studies have demonstrated a J-shaped association between alcohol consumption and diseases such as coronary heart disease, diabetes, and hypertension, with lower prevalence in light to moderate drinkers compared with nondrinkers and the highest prevalence in heavy drinkers.18 Another study demonstrated a similar J-shaped association between alcohol consumption and CRP levels.19 Our study did not have detailed information on higher categories of alcohol intake, as the study questionnaire had prespecified fields for 0, 1, 2, or 3 or more drinks per day. Also, we do not know whether the protective association observed in the current study is with alcohol or with a healthy lifestyle that includes alcohol. Alternatively, it could be a marker of unmeasured risk factors associated with individuals who abstain from alcohol.
Strengths of the study include its large sample size, high participation rate, and the availability of biomarkers not often available in studies of this size. A fasting state in participants included in the analysis allowed proper interpretation of glucose levels and HOMA-IR equation–derived insulin resistance. Although the study had several strengths, there are also limitations to this study methodology. Due to the cross-sectional nature of the study, causality cannot be determined. In addition, several measures such as CVD and hypertension case identification were based on self-report and are therefore subject to recall bias and misclassification. There was no measure of waist circumference, which is thought to be a better measure of central adiposity than BMI. However, in the study by Wildman and colleagues,5 similar results were obtained when weight status was determined using waist circumference rather than BMI. Another limitation is that only 15,000 study participants were available with >8 hours fasting status and therefore eligible to be included in the current analysis. We found that those who did not fast and were therefore excluded had, on average, a healthier lifestyle profile than those who were included. Therefore, it is possible that the proportion of participants with obesity and cardiometabolic risk factor clustering that we report in the current study may be an overestimate.
In summary, we found that one-fourth of normal-weight adults displayed cardiometabolic risk factor clustering, while one-fourth of obese adults did not display clustering of cardiometabolic risk factors. These findings suggest that within the traditionally used BMI categories, there is a heterogeneous pattern of cardiometabolic risk factor clustering that has an independent effect on cardiovascular risk. When compared with normal-weight subjects without cardiometabolic risk factor clustering, we found that overweight/obese individuals without cardiometabolic risk factor clustering may not have a higher CVD risk; in contrast, normal-weight individuals who displayed clustering of cardiometabolic risk factors had more than twice the risk of CVD. A corollary observation to our findings is that individual-level as well as public health interventions that focus on reducing weight alone, without addressing the metabolic profile, may not adequately address the risk of CVD. Given our findings, screening for metabolic factors recommended in the American Heart Association prevention guidelines, such as lipids, blood pressure, CRP, and others should be considered in normal-weight adults in addition to overweight and obese adults.20