SAN ANTONIO – Excess visceral fat at baseline, subsequent weight gain, and markers of insulin resistance were among factors independently associated with incident prediabetes and type 2 diabetes in obese adults in the Dallas Heart Study.
General adiposity was not found to be associated with increased prediabetes and type 2 diabetes risk, Dr. James A. de Lemos reported at the annual meeting of the Obesity Society.
The findings provide insight as to why some obese individuals develop diabetes while others do not, said Dr. de Lemos of the University of Texas Southwestern Medical Center, Dallas.
The epidemic of obesity has led to a secondary epidemic in diabetes that has offset any improvements that would have occurred to the cardiovascular event rate due to lower rates of smoking, lower levels of low-density lipoprotein cholesterol, and less hypertension in the population, Dr. de Lemos explained. Diabetes manifestations are heterogeneous among obese individuals, and body mass index doesn’t help discriminate individuals who are at risk for diabetes versus those who aren’t, he added.
"Our hypothesis was that markers of adipose tissue dysfunction may better predict diabetes and prediabetes onset than general adiposity markers in a specifically obese population," he said.
Of 732 participants in the multiethnic population-based cohort study who were obese at baseline, 84 (11.5%) developed diabetes.
Independent predictors of diabetes on multivariate analysis included higher baseline visceral fat mass (odds ratio per 1 standard deviation 1.4 kg, 2.4), fructosamine level (1.1 micromol/L, 2.0), fasting glucose level (1.1 micromol/L, 1.9), family history of diabetes (OR 2.3), systolic blood pressure (OR per 10 mm Hg, 1.3), and weight gain over follow-up (OR per 1 kg, 1.06). No associations were noted for body mass index, total body fat, or abdominal subcutaneous fat, Dr. de Lemos and his colleagues found (JAMA. 2012; 308: 1150-9).
Conversely, lower body fat mass and adiponectin level showed significant graded, inverse associations with incident prediabetes and diabetes.
Among 512 study participants with normal baseline glucose values, 39% experienced a composite outcome of prediabetes or diabetes. Factors significantly associated with this outcome included baseline measurements of visceral fat mass, fasting glucose level, insulin, and fructosamine, as well as older age, nonwhite race, family history of diabetes, and weight gain over follow-up, Dr. de Lemos said, noting that this composite outcome also was not associated with measures of general adiposity.
Notably, weight gain in this subgroup was the most powerful predictor of prediabetes and diabetes risk over the follow-up period, Dr. de Lemos said.
Also, an evaluation of measures of subclinical cardiovascular disease showed that those who went on to develop prediabetes or diabetes had a greater prevalence of coronary calcium and left ventricular hypertrophy well before onset of these conditions.
"This suggests that some of the factors that contribute to the development of prediabetes and diabetes may be contributing to the development of subclinical cardiovascular disease even before prediabetes and diabetes are manifest," he said.
The Dallas Heart Study enrolled more than 6,100 patients, with oversampling of African Americans, who made up half of the study population. The subset of participants included in the current analysis were adults aged 30-65 years with a BMI of 30 or higher who were enrolled between 2000 and 2002 and followed for a median of 7 years. Body composition was measured using dual energy x-ray absorptiometry scanning and magnetic resonance imaging; subclinical atherosclerosis and cardiac structure and function were measured by computed tomography and MRI. Circulating adipokines and biomarkers of insulin resistance, dyslipidemia, and inflammation also were measured.
The findings indicate that a dysfunctional adiposity phenotype, characterized by excess visceral fat mass and insulin resistance, is associated with both incident prediabetes and diabetes in a specifically obese population, Dr. de Lemos said. This, along with the absence of an association between markers of general adiposity and incident prediabetes and diabetes, suggests that obesity is a "heterogeneous disorder with a distinct adiposity set of phenotypes.
"Important is the fact that many of our obese participants over this 8-year period did not develop diabetes or prediabetes, and it remains likely that some favorable metabolic phenotypes exist within the obese population," he added.
Although caution must be taken in drawing direct clinical implications with respect to therapy from an observational study such as this, it is reasonable to consider the possibility that incorporating measures of visceral fat mass and biomarkers of insulin resistance may help to characterize the risk of diabetes among obese individuals, and might be used to identify individuals who would most benefit from intensive lifestyle therapy, he said.