TO012 DYSFUNCTIONAL HIGH-DENSITY LIPOPROTEIN (HDL) IN CHRONIC KIDNEY DISEASE (CKD): A PROSPECTIVE OBSERVATIONALTRIALOF 3.306 PATIENTS

Introduction and Aims: In healthy subjects, HDL-cholesterol (HDL) serum levels are inversely correlated with the rate of cardiovascular events. Recent studies revealed that under uremic conditions the vasoprotective properties of HDL may be altered by incorporation of serum amyloid A (SAA) in the HDL particle, resulting in reduced endothelial nitric oxide (NO) and an enhanced superoxide production. However, the clinical relevance of these findings remains unclear. We assessed the relationship between HDL levels and all-cause or cardiovascular mortality in a large cohort of patients undergoing coronary angiography. Methods: The LURIC study enrolled 3.306 patients between 1997 and 2000 undergoing coronary angiography. Patients were followed for median of 9.9 years. HDL and SAA levels were measured at inclusion. To assess the effect of HDL and SAA on all-cause and cardiovascular mortality, Cox regression analyses were performed. Patients were divided for SAA or HDL quartiles, respectively. Additionally, we determined the effect of increasing HDL concentrations on endothelial NO and superoxide production in presence of SAA by using electron-spin resonance spectroscopy. Results: Patients with CKD (cystatin C eGFR ≤60 ml/min, n=474) exhibited significantly reduced HDL levels, while SAA levels were increased as compared to patients without CKD (n=2.832). In the whole study population, we identified increasing SAA levels as strong predictor for all-cause and cardiovascular mortality during follow-up. Interestingly, high HDL levels compensated for the adverse effect of SAA on mortality. Indeed, we confirmed that HDL ameliorated the effect of SAA on endothelial NO and superoxide production in-vitro. However, this compensatory effect of HDL was dramatically reduced in CKD patients, indicating a loss of the vasoprotective effects of HDL. Therefore, we determined the effect of high HDL levels per se on mortality. High HDL levels significantly reduced the risk for all-cause (HR 0.743, p=0.012) and cardiovascular (HR 0.683, p=0.013) mortality in patients without CKD. In marked contrast, higher levels of HDL did not significantly reduce the risk for all-cause and cardiovascular mortality during follow-up in patients with eGFR ≤60 ml/ min (HR 1.003, p=0.986 and HR 0.925, p=0.705). Conclusions: These results show for the first time that HDL loses its vasoprotective properties in patients with CKD compared to those without CKD. Moreover, we were able to demonstrate that higher HDL levels do not reduce the risk for all-cause and cardiovascular mortality in CKD patients. These findings underscore the relevance of HDL as an important factor in pathogenesis of atherosclerotic disease in patients with CKD.