Abstract
The aim of this study was to investigate associations of cardiovascular outcomes with lipid variability across body mass index categories. We identified 6689 participants (57.1% women) enrolled in the Atherosclerosis Risk In Communities (ARICs) study who had ≥ 3 measurements of total cholesterol (TC) and low-density lipoprotein cholesterol (LDL-C). Cox regression models were used to compute hazard ratios (HRs)-associated heart failure (HF), myocardial infarction (MI), and mortality with 1-SD increase in lipid variability captured by SD and variability independent of the mean (VIM). We also assessed whether adding lipid variability would improve the cardiovascular risk prediction beyond the conventional risk factors. Among 2130 (31.8%) obese patients, 1907 (89.5%) had obesity classes I and II and 223 (10.5%) had obesity class III. In multivariable-adjusted analyses, TC and LDL-C variabilities were significantly (p ≤ 0.047) associated with HF in overweight (HRs ranging from 1.10 to 1.17), obesity classes I and II (1.11-1.14), and obesity class III (1.21-1.39). Higher TC and LDL-C variabilities conferred higher risk of MI and mortality in obesity classes I and II (p ≤ 0.007). Adding TC-VIM and LDL-C-VIM rather than the lipid level to a conventional risk model significantly improved risk prediction of HF with net reclassification improvement amounting to 8.95% for TC-VIM (p=0.006) and 8.09% for LDL-C-VIM (p=0.012). Elevated TC and LDL-C variabilities were associated with the increased risk of cardiovascular outcomes, particularly in obesity. Our observations highlight the importance of lipid variability in obesity-associated dyslipidemia.