Abstract 84: Serum Amyloid A Does Not Modulate the Inverse Association of HDL with Coronary Artery Calcification in Type 2 Diabetes

Abstract

Introduction Although high-density lipoprotein (HDL) is inversely correlated with cardiovascular risk, HDL loses its protective role in pathologic inflammatory states like type 2 diabetes (T2DM). HDL dysfunction contributes to accelerated atherosclerosis in T2DM, but the mechanism is incompletely defined. The acute phase reactant serum amyloid A (SAA) displaces apolipoprotein A-I and may impair HDL-mediated reverse cholesterol efflux. We hypothesized that SAA alters the inverse association between HDL and coronary artery calcium (CAC) in the Penn Diabetes Heart Study, a cross-sectional study of T2DM patients free of overt cardiovascular or renal disease. Methods We measured SAA in serum samples by immunonephelometry (N=975; mean age 58 ± 9 years; 63% male, 57% Caucasian; mean BMI 33 ± 6 kg/m 2 ). HDL was measured enzymatically in lipoprotein fractions after ultracentrifugation. Agatston CAC scores were quantified from electron beam tomography at the same visit. Spearman correlation and logistic regression were used to test associations of SAA with clinical factors and metabolic syndrome. We used Tobit regression to analyze associations between CAC and HDL, both overall and stratified by 3 categories of SAA: undetectable, lower half detectable, and upper half detectable. Results Spearman correlations revealed moderate association of SAA with C-reactive protein (r=0.52) and weak associations of SAA with BMI (r=0.25) and HDL (r=0.17; all p<0.001). In logistic regression, the group with highest SAA levels had increased odds of metabolic syndrome compared to those with undetectable levels (OR 1.56, 95% CI 1.03 to 2.38, p=0.036). In adjusted Tobit regression, HDL was inversely associated with CAC (Tobit coefficient for 1-SD increase in HDL: -0.30; 95% CI -0.54 to -0.06; p=0.013). Across the categories of SAA, however, there was no difference in the association of HDL with CAC (Tobit coefficient for 1-SD increase in HDL: -0.17 [95% CI -0.49 to 0.16] for undetectable vs. -0.31 [95% CI -0.79 to 0.17] for lower half detectable vs. -0.49 [95% CI -1.01 to 0.03] for upper half detectable). Conclusions Despite the association of SAA with metabolic syndrome, these data suggest that elevated SAA may not change the inverse relationship of HDL with CAC in T2DM.