The U-Shaped Association of Non-High-Density Lipoprotein Cholesterol Levels With All-Cause and Cardiovascular Mortality Among Patients With Hypertension

Abstract

Background: Non-high-density lipoprotein cholesterol (non-HDL-C) is a valuable indicator in routine blood lipid tests, but the associations of non-HDL-C with mortality in hypertensive population still remain uncertain.Methods: In the National Health and Nutrition Examination Surveys from 1999 to 2014, participants having hypertension were included and grouped by non-HDL-C levels (<130, 130–159, 160–189, 190–219, and ≥220 mg/dl). Multivariate Cox regression was conducted for calculation of hazard ratios (HR) and 95% confidence interval (CI). To reveal the relationship between non-HDL-C and mortality, Kaplan–Meier survival curves, restricted cubic spline, linear regression, and subgroup analysis were also applied.Results: A total of 12,169 participants (47.52% males, mean age 57.27 ± 15.79 years) were included. During average follow-up of 92.5 months, 1,946 (15.99%) all-cause deaths and 422 (3.47%) cardiovascular deaths occurred. After adjusting for confounders, the association of non-HDL-C with mortality was detected as U-shaped. Threshold values were observed at 158 mg/dl for all-cause mortality and 190 mg/dl as to cardiovascular mortality. Below the threshold, every 10 mg/dl increment in non-HDL-C attributed to relatively low all-cause mortality significantly (HR = 0.94, 95% CI: 0.92–0.96). Above the threshold, non-HDL-C has significant positive associations with both all-cause (HR = 1.03, 95% CI: 1.01–1.05) and cardiovascular mortality (HR = 1.09, 95% CI: 1.05–1.14). For subgroups analysis, similar results were found among participants age <65 years old, non-white population, those were not taking lipid-lowering drugs, and subjects with body mass index (BMI) ≥25 kg/m2.Conclusion: The U-shaped association was detected between non-HDL-C and mortality among hypertensive population.

U-Shaped Relationship of Non-HDL Cholesterol With All-Cause and Cardiovascular Mortality in Men Without Statin Therapy

“…Moreover, the present study contributed evidence that lower non-HDL-C is also closely related to higher mortality in men without statin therapy, and indicated a U-shaped association. Although the disparity in the study population, in accordance with our results, several previous studies have observed the U-shaped association between non-HDL-C and mortality ( 11 , 12 ). One study by Cheng and colleagues analysing data from NHANES demonstrated that relatively higher or lower non-HDL-C concentrations were linked to increased mortality, and the lowest risk was found at threshold values of 158 and 190 mg/dl for all-cause and cardiovascular mortality, respectively.…”

“…A systematic review and meta-analysis by our team previously demonstrated that the increased levels of non-HDL-C were significantly associated with an increased risk of mortality in patients with coronary heart disease (CHD) (8), which was similar to previous research findings in population without cardiovascular diseases (9) or patients with diabetes (10). Interestingly, a U-shaped relationship was also recently identified in different populations (11,12). Studies detected the U-shaped association of non-HDL-C with all-cause and cardiovascular mortality among patients with hypertension or chronic kidney disease (CKD) stages 3-5 (11,12), non-HDL-C was U-shaped associated with mortality among male hypertension in subgroup analysis, but not in female (12).…”

“…Interestingly, a U-shaped relationship was also recently identified in different populations (11,12). Studies detected the U-shaped association of non-HDL-C with all-cause and cardiovascular mortality among patients with hypertension or chronic kidney disease (CKD) stages 3-5 (11,12), non-HDL-C was U-shaped associated with mortality among male hypertension in subgroup analysis, but not in female (12). Differently, one study focused on the men population, from the Israeli National Death Registry demonstrated a positive association that non-HDL-C as a useful predictor of cardiovascular disease mortality in 22 years followup (13).…”

“…First, one possible reason is that the participants with the lowest cholesterol levels had a poorer health status ( 28 ), or debilitation and illness have been hypothesised to cause a decrease in levels of cholesterol ( 26 , 30 , 31 ). Second, higher HDL-C equals to low non-HDL-C levels according to the calculation formula, extremely high HDL-C increases mortality in the general population by analysing the data from NHANES ( 12 ). The genetic variation of particular genes and variation of the size or function of HDL particles may be the underlying mechanism ( 12 ).…”

“…Second, higher HDL-C equals to low non-HDL-C levels according to the calculation formula, extremely high HDL-C increases mortality in the general population by analysing the data from NHANES ( 12 ). The genetic variation of particular genes and variation of the size or function of HDL particles may be the underlying mechanism ( 12 ). Finally, the U-shaped association between lipoprotein levels and mortality may be similar to the obesity paradox, which is largely explained by methodological issues, including reverse causation.…”