Abstract
OBJECTIVE: Reasons for variations in atherosclerotic burden among individuals with similar levels of obesity are poorly understood, especially in African Americans. This study examines whether high-sensitivity C-reactive protein (hsCRP) is useful for discriminating between benign and high-risk obesity phenotypes for subclinical atherosclerosis in African Americans. APPROACH AND RESULTS: Participants from the Jackson Heart Study (n=4682) were stratified into 4 phenotypes based on the presence of National Heart and Lung and Blood Institute definition of obesity or obesity-equivalent (body mass index ≥30 or body mass index 25-30 with waist circumference >102 cm in men and >88 cm in women) and inflammation by hsCRP ≥2 mg/L. Using multivariate regression models, we conducted cross-sectional analyses of the association between inflammatory obesity phenotypes and subclinical atherosclerosis determined by carotid intima-media thickness or coronary artery calcium scores. Sex-specific analyses were conducted given significant interaction for gender (P=0.03). The prevalence of obesity or equivalent was 65%, of which 30% did not have inflammation. Conversely, 37% of nonobese individuals had inflammation. Among nonobese men, hsCRP ≥2 mg/L identified a subset of individuals with higher carotid intima-media thickness (adjusted mean difference =0.05, 95% confidence interval 0.02, 0.08 mm) compared with their noninflammatory counterparts. Among obese men, hsCRP <2 mg/L identified a subset of individuals with lower coronary artery calcium compared with their inflammatory counterparts. Among women, associations between hsCRP and carotid intima-media thickness or coronary artery calcium were not found. CONCLUSIONS: In the largest African American population-based cohort to date, hsCRP was useful in identifying a subset of nonobese men with higher carotid intima-media thickness, but not in women. hsCRP did not identify a subset of obese individuals with less subclinical atherosclerosis.