Chronic kidney disease, lipids and apolipoproteins, and coronary heart disease: The ARIC Study

BACKGROUND: Chronic kidney disease (CKD) is associated with elevated apolipoprotein B to A-1 ratio (ApoB/A1). It is not known whether these markers are more strongly associated with the risk of coronary heart disease (CHD) in CKD compared to traditionally measured lipids and lipoprotein cholesterol ratios. METHODS: We studied the association of lipids and apolipoproteins including non-HDL-cholesterol to HDL-cholesterol ratio (NonHDLc/HDLc) and ApoB/A1 with incident CHD in 10,137 individuals free of CHD at baseline (visit four) in the Atherosclerosis Risk in Communities (ARIC) study. An estimated glomerular filtration rate of 15 to <60 ml/min/1.73 m(2) based on a cystatin C measurement was used to define CKD (Stage 3-4). Cox proportional hazards regression models were used to determine the association of lipids and apolipoprotein measurements with the risk of CHD in those with and without CKD after adjustment for demographic and known clinical cardiovascular risk factors. RESULTS: CKD was present in 1217 (12%) individuals free of CHD at baseline. The median follow-up time was 11.1 years. A CHD event developed in 498 out of 8920 individuals without CKD (incidence rate: 5.2 events per 1000 person-years) and in 138 out of 1217 individuals with CKD (incidence rate: 12.0 events per 1000 person-years; P < 0.001). Those with CKD had a lower concentration of ApoA1: median (in g/L) and interquartile range (IQR) = 1.40 (1.38-1.42) vs. 1.48 (1.47-1.49) P < 0.001; and a higher ApoB/A1 = 0.75 (0.73-0.77) vs. 0.71 (0.70-0.72) P < 0.001; than those without CKD (eGFR >= 60 ml/min/1.73 m(2)). Among individuals with CKD, ApoB/A1 and NonHDLc/HDLc were both associated with the risk of CHD: hazard ratios (HR) and 95% confidence intervals (CI) per one standard deviation increase = 1.22 (1.02-1.46) for ApoB/A1 and 1.30 (1.07-1.57) for NonHDLc/HDLc with no significant differences detected (P for interaction >0.1) when comparing these estimates to those of participants without CKD. CONCLUSIONS: Although CKD is associated with a lower ApoA1 concentration and with a higher ApoB/A1, we found no evidence that these apolipoproteins are more strongly associated with CHD incidence in CKD compared to NonHDLc/HDLc. (C) 2014 Elsevier Ireland Ltd. All rights reserved.