Abstract
INTRODUCTION: Abnormalities in total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), and triglycerides (TG) are each associated with increased cardiovascular risk, after adjusting for non-lipid risk factors. However, whether and to what extent the association for each lipid measure is confounded by other lipid measures is less understood. AIM: This study aims to investigate the association of each lipid measure with cardiovascular and all-cause mortality while precluding the confounding caused by abnormalities in other lipid measures. METHODS: The study utilized data from the third National Health and Nutrition Examination Survey (NHANES III, 1988-1994) and ten cycles of continuous NHANES (1999-2018). The study cohort included 23,761 participants who were 20 years or older, not pregnant, not receiving lipid-lowering treatment, and had complete data on all four lipid measures and mortality status. Participants were categorized into seven subgroups based on their lipid profiles. Kaplan-Meier survival curves and Cox proportional hazards models were used to examine the association between lipid abnormalities and mortality. RESULTS: During a median follow-up of 140 months, 5,003 participants (14.1%) died, with 1,665 deaths (4.2%) attributable to cardiovascular causes. Compared with the reference group in which the four lipid measures were all normal, the subgroups with isolated high TC, two to three lipid abnormalities, and four lipid abnormalities were associated with increased risks for both cardiovascular and all-cause mortality in univariate analysis. However, only those with isolated high TC (for cardiovascular mortality, HR 1.52, 95% CI 1.13-2.06) and four lipid abnormalities (for all-cause mortality, HR 1.34, 95% CI 1.04-1.72) remained statistically significant after adjusting for non-lipid risk factors. Of note, compared with the reference group, the profile of non-lipid risk factors was apparently less favorable in the subgroup with two to three lipid abnormalities but similar (and some factors even more favorable) in the subgroup with isolated high TC. When the lipid measures were analyzed as continuous variables, a U-shaped relationship between HDL-C and mortality risk was observed for both cardiovascular and all-cause mortality, and very low LDL-C level was associated with increased mortality risk. No statistically significant association was found between TG levels and mortality risk. CONCLUSION: Isolated high TC, very low LDL-C, and concurrent abnormalities in all four lipid measures were associated with increased mortality risk, whereas isolated high TG was not. A U-shaped relationship may exist between HDL-C level and mortality. Overall, these findings underscore the need for integrated management of dyslipidemia that takes all four lipid measures as well as non-lipid cardiovascular risk factors into account, particularly for those with concurrent abnormalities in two or more lipid measures.